Search Results for author:
Found 276 papers, 144 papers with code
Continual Learning via Explicit Structure Learning
During structure learning, the model optimizes for the best structure for the current task.
Are Pre-trained Transformers Robust in Intent Classification? A Missing Ingredient in Evaluation of Out-of-Scope Intent Detection
Pre-trained Transformer-based models were reported to be robust in intent classification.
Few-Shot Intent Classification by Gauging Entailment Relationship Between Utterance and Semantic Label
Compared with DNNC, our proposed method is more efficient in both training and serving since it is based upon the entailment between query utterance and labels instead of all the training examples.
The Thieves on Sesame Street are Polyglots - Extracting Multilingual Models from Monolingual APIs
Pre-training in natural language processing makes it easier for an adversary with only query access to a victim model to reconstruct a local copy of the victim by training with gibberish input data paired with the victim{'}s labels for that data.
Simple Data Augmentation with the Mask Token Improves Domain Adaptation for Dialog Act Tagging
The concept of Dialogue Act (DA) is universal across different task-oriented dialogue domains - the act of {``}request{''} carries the same speaker intention whether it is for restaurant reservation or flight booking.
[CASPI] Causal-aware Safe Policy Improvement for Task-oriented Dialogue
Further more we demonstrate sample efficiency, where our method trained only on 20% of the data, are comparable to current state of the art method trained on 100% data on two out of there evaluation metrics.
DocQueryNet: Value Retrieval with Arbitrary Queries for Form-like Documents
Unlike previous methods that only address a fixed set of field items, our method predicts target value for an arbitrary query based on the understanding of the layout and semantics of a form.
OSWorld: Benchmarking Multimodal Agents for Open-Ended Tasks in Real Computer Environments
Autonomous agents that accomplish complex computer tasks with minimal human interventions have the potential to transform human-computer interaction, significantly enhancing accessibility and productivity.
What Are We Measuring When We Evaluate Large Vision-Language Models? An Analysis of Latent Factors and Biases
Vision-language (VL) models, pretrained on colossal image-text datasets, have attained broad VL competence that is difficult to evaluate.
How Much are LLMs Contaminated? A Comprehensive Survey and the LLMSanitize Library
With the rise of Large Language Models (LLMs) in recent years, new opportunities are emerging, but also new challenges, and contamination is quickly becoming critical.
FOFO: A Benchmark to Evaluate LLMs' Format-Following Capability
This paper presents FoFo, a pioneering benchmark for evaluating large language models' (LLMs) ability to follow complex, domain-specific formats, a crucial yet underexamined capability for their application as AI agents.
AgentOhana: Design Unified Data and Training Pipeline for Effective Agent Learning
It meticulously standardizes and unifies these trajectories into a consistent format, streamlining the creation of a generic data loader optimized for agent training.
AgentLite: A Lightweight Library for Building and Advancing Task-Oriented LLM Agent System
Thus, we open-source a new AI agent library, AgentLite, which simplifies this process by offering a lightweight, user-friendly platform for innovating LLM agent reasoning, architectures, and applications with ease.
Unified Training of Universal Time Series Forecasting Transformers
Deep learning for time series forecasting has traditionally operated within a one-model-per-dataset framework, limiting its potential to leverage the game-changing impact of large pre-trained models.
Causal Layering via Conditional Entropy
Under appropriate assumptions and conditioning, we can separate the sources or sinks from the remainder of the nodes by comparing their conditional entropy to the unconditional entropy of their noise.
Editing Arbitrary Propositions in LLMs without Subject Labels
On datasets of binary propositions derived from the CounterFact dataset, we show that our method -- without access to subject labels -- performs close to state-of-the-art L\&E methods which has access subject labels.
Parameter-Efficient Detoxification with Contrastive Decoding
DETOXIGEN is an ensemble of a pre-trained language model (generator) and a detoxifier.
TrustLLM: Trustworthiness in Large Language Models
This paper introduces TrustLLM, a comprehensive study of trustworthiness in LLMs, including principles for different dimensions of trustworthiness, established benchmark, evaluation, and analysis of trustworthiness for mainstream LLMs, and discussion of open challenges and future directions.
Moonshot: Towards Controllable Video Generation and Editing with Multimodal Conditions
This work presents Moonshot, a new video generation model that conditions simultaneously on multimodal inputs of image and text.
Unlocking Anticipatory Text Generation: A Constrained Approach for Large Language Models Decoding
Large Language Models (LLMs) have demonstrated a powerful ability for text generation.
X-InstructBLIP: A Framework for aligning X-Modal instruction-aware representations to LLMs and Emergent Cross-modal Reasoning
Vision-language pre-training and instruction tuning have demonstrated general-purpose capabilities in 2D visual reasoning tasks by aligning visual encoders with state-of-the-art large language models (LLMs).
ChatGPT's One-year Anniversary: Are Open-Source Large Language Models Catching up?
Upon its release in late 2022, ChatGPT has brought a seismic shift in the entire landscape of AI, both in research and commerce.
Diffusion Model Alignment Using Direct Preference Optimization
Large language models (LLMs) are fine-tuned using human comparison data with Reinforcement Learning from Human Feedback (RLHF) methods to make them better aligned with users' preferences.
Lexical Repetitions Lead to Rote Learning: Unveiling the Impact of Lexical Overlap in Train and Test Reference Summaries
Ideal summarization models should generalize to novel summary-worthy content without remembering reference training summaries by rote.
Fair Abstractive Summarization of Diverse Perspectives
However, current work in summarization metrics and Large Language Models (LLMs) evaluation has not explored fair abstractive summarization.
Are You Sure? Challenging LLMs Leads to Performance Drops in The FlipFlop Experiment
The interactive nature of Large Language Models (LLMs) theoretically allows models to refine and improve their answers, yet systematic analysis of the multi-turn behavior of LLMs remains limited.
Salespeople vs SalesBot: Exploring the Role of Educational Value in Conversational Recommender Systems
Making big purchases requires consumers to research or consult a salesperson to gain domain expertise.
OpenAgents: An Open Platform for Language Agents in the Wild
Language agents show potential in being capable of utilizing natural language for varied and intricate tasks in diverse environments, particularly when built upon large language models (LLMs).
How Do Transformers Learn In-Context Beyond Simple Functions? A Case Study on Learning with Representations
Through extensive probing and a new pasting experiment, we further reveal several mechanisms within the trained transformers, such as concrete copying behaviors on both the inputs and the representations, linear ICL capability of the upper layers alone, and a post-ICL representation selection mechanism in a harder mixture setting.
Lemur: Harmonizing Natural Language and Code for Language Agents
We introduce Lemur and Lemur-Chat, openly accessible language models optimized for both natural language and coding capabilities to serve as the backbone of versatile language agents.
L2CEval: Evaluating Language-to-Code Generation Capabilities of Large Language Models
Recently, large language models (LLMs), especially those that are pretrained on code, have demonstrated strong capabilities in generating programs from natural language inputs in a few-shot or even zero-shot manner.
Beyond the Chat: Executable and Verifiable Text-Editing with LLMs
Conversational interfaces powered by Large Language Models (LLMs) have recently become a popular way to obtain feedback during document editing.
Embrace Divergence for Richer Insights: A Multi-document Summarization Benchmark and a Case Study on Summarizing Diverse Information from News Articles
In this paper, we propose a new task of summarizing diverse information encountered in multiple news articles encompassing the same event.
XGen-7B Technical Report
Most open-source LLMs, on the other hand, are limited in their ability to support longer sequence lengths, which is a key requirement for many tasks that require inference over an input context.
Modeling Uncertainty and Using Post-fusion as Fallback Improves Retrieval Augmented Generation with LLMs
This paper aims to fill this gap by investigating different methods of combining retrieved passages with LLMs to enhance answer generation.
Enhancing Performance on Seen and Unseen Dialogue Scenarios using Retrieval-Augmented End-to-End Task-Oriented System
End-to-end task-oriented dialogue (TOD) systems have achieved promising performance by leveraging sophisticated natural language understanding and natural language generation capabilities of pre-trained models.
BOLAA: Benchmarking and Orchestrating LLM-augmented Autonomous Agents
The massive successes of large language models (LLMs) encourage the emerging exploration of LLM-augmented Autonomous Agents (LAAs).
Retroformer: Retrospective Large Language Agents with Policy Gradient Optimization
This demonstrates that using policy gradient optimization to improve language agents, for which we believe our work is one of the first, seems promising and can be applied to optimize other models in the agent architecture to enhance agent performances over time.
DialogStudio: Towards Richest and Most Diverse Unified Dataset Collection for Conversational AI
Despite advancements in conversational AI, language models encounter challenges to handle diverse conversational tasks, and existing dialogue dataset collections often lack diversity and comprehensiveness.
REX: Rapid Exploration and eXploitation for AI Agents
In this paper, we propose an enhanced approach for Rapid Exploration and eXploitation for AI Agents called REX.
Sample-Efficient Learning of POMDPs with Multiple Observations In Hindsight
This paper studies the sample-efficiency of learning in Partially Observable Markov Decision Processes (POMDPs), a challenging problem in reinforcement learning that is known to be exponentially hard in the worst-case.
Did You Read the Instructions? Rethinking the Effectiveness of Task Definitions in Instruction Learning
Large language models (LLMs) have shown impressive performance in following natural language instructions to solve unseen tasks.
SWiPE: A Dataset for Document-Level Simplification of Wikipedia Pages
Text simplification research has mostly focused on sentence-level simplification, even though many desirable edits - such as adding relevant background information or reordering content - may require document-level context.
LLMs as Factual Reasoners: Insights from Existing Benchmarks and Beyond
To address this, we propose a new protocol for inconsistency detection benchmark creation and implement it in a 10-domain benchmark called SummEdits.
UniControl: A Unified Diffusion Model for Controllable Visual Generation In the Wild
Visual generative foundation models such as Stable Diffusion show promise in navigating these goals, especially when prompted with arbitrary languages.
ULIP-2: Towards Scalable Multimodal Pre-training for 3D Understanding
Recent advancements in multimodal pre-training methods have shown promising efficacy in 3D representation learning by aligning multimodal features across 3D shapes, their 2D counterparts, and language descriptions.
Ranked #6 on
3D Point Cloud Classification
on ScanObjectNN
(using extra training data)
HPE:Answering Complex Questions over Text by Hybrid Question Parsing and Execution
It comprises two central pillars: (1) We parse the question of varying complexity into an intermediate representation, named H-expression, which is composed of simple questions as the primitives and symbolic operations representing the relationships among them; (2) To execute the resulting H-expressions, we design a hybrid executor, which integrates the deterministic rules to translate the symbolic operations with a drop-in neural reader network to answer each decomposed simple question.
Zero-shot Item-based Recommendation via Multi-task Product Knowledge Graph Pre-Training
This paper presents a novel paradigm for the Zero-Shot Item-based Recommendation (ZSIR) task, which pre-trains a model on product knowledge graph (PKG) to refine the item features from PLMs.
CodeGen2: Lessons for Training LLMs on Programming and Natural Languages
In this study, we attempt to render the training of LLMs for program synthesis more efficient by unifying four key components: (1) model architectures, (2) learning methods, (3) infill sampling, and, (4) data distributions.
Efficiently Aligned Cross-Lingual Transfer Learning for Conversational Tasks using Prompt-Tuning
Our results demonstrate the strong and efficient modeling ability of NLI-based classifiers and the large cross-lingual transfer improvements achieved by our aligned prompts, particularly in few-shot settings.
GlueGen: Plug and Play Multi-modal Encoders for X-to-image Generation
Empirical results show that GlueNet can be trained efficiently and enables various capabilities beyond previous state-of-the-art models: 1) multilingual language models such as XLM-Roberta can be aligned with existing T2I models, allowing for the generation of high-quality images from captions beyond English; 2) GlueNet can align multi-modal encoders such as AudioCLIP with the Stable Diffusion model, enabling sound-to-image generation; 3) it can also upgrade the current text encoder of the latent diffusion model for challenging case generation.
HIVE: Harnessing Human Feedback for Instructional Visual Editing
Incorporating human feedback has been shown to be crucial to align text generated by large language models to human preferences.
On the Unlikelihood of D-Separation
We then provide an analytic average case analysis of the PC Algorithm for causal discovery, as well as a variant of the SGS Algorithm we call UniformSGS.
Towards Interpretable and Efficient Automatic Reference-Based Summarization Evaluation
Interpretability and efficiency are two important considerations for the adoption of neural automatic metrics.
Fantastic Rewards and How to Tame Them: A Case Study on Reward Learning for Task-oriented Dialogue Systems
Prior works mainly focus on adopting advanced RL techniques to train the ToD agents, while the design of the reward function is not well studied.
A Comprehensive Survey on Pretrained Foundation Models: A History from BERT to ChatGPT
This study provides a comprehensive review of recent research advancements, challenges, and opportunities for PFMs in text, image, graph, as well as other data modalities.
Designing and Evaluating Interfaces that Highlight News Coverage Diversity Using Discord Questions
In a second usability study, we developed and implemented a reading exercise with 95 novice news readers to measure exposure to coverage diversity.
Improved Online Conformal Prediction via Strongly Adaptive Online Learning
We prove that our methods achieve near-optimal strongly adaptive regret for all interval lengths simultaneously, and approximately valid coverage.
Lower Bounds for Learning in Revealing POMDPs
However, the fundamental limits for learning in revealing POMDPs are much less understood, with existing lower bounds being rather preliminary and having substantial gaps from the current best upper bounds.
Salesforce CausalAI Library: A Fast and Scalable Framework for Causal Analysis of Time Series and Tabular Data
Finally, we provide a user interface (UI) that allows users to perform causal analysis on data without coding.
Model-Agnostic Hierarchical Attention for 3D Object Detection
By plugging our proposed modules into the state-of-the-art transformer-based 3D detector, we improve the previous best results on both benchmarks, with the largest improvement margin on small objects.
LayoutDETR: Detection Transformer Is a Good Multimodal Layout Designer
Graphic layout designs play an essential role in visual communication.
Revisiting the Gold Standard: Grounding Summarization Evaluation with Robust Human Evaluation
Human evaluation is the foundation upon which the evaluation of both summarization systems and automatic metrics rests.
ULIP: Learning a Unified Representation of Language, Images, and Point Clouds for 3D Understanding
Then, ULIP learns a 3D representation space aligned with the common image-text space, using a small number of automatically synthesized triplets.
Ranked #3 on
Training-free 3D Point Cloud Classification
on ModelNet40
(using extra training data)
Best-$k$ Search Algorithm for Neural Text Generation
We first investigate the vanilla best-first search (BFS) algorithm and then propose the Best-$k$ Search algorithm.
SPE: Symmetrical Prompt Enhancement for Fact Probing
In this work, we propose Symmetrical Prompt Enhancement (SPE), a continuous prompt-based method for factual probing in PLMs that leverages the symmetry of the task by constructing symmetrical prompts for subject and object prediction.
Improving Factual Consistency in Summarization with Compression-Based Post-Editing
We propose to use sentence-compression data to train the post-editing model to take a summary with extrinsic entity errors marked with special tokens and output a compressed, well-formed summary with those errors removed.
Uni-Parser: Unified Semantic Parser for Question Answering on Knowledge Base and Database
Parsing natural language questions into executable logical forms is a useful and interpretable way to perform question answering on structured data such as knowledge bases (KB) or databases (DB).
Discord Questions: A Computational Approach To Diversity Analysis in News Coverage
There are many potential benefits to news readers accessing diverse sources.
Model ensemble instead of prompt fusion: a sample-specific knowledge transfer method for few-shot prompt tuning
Through this way, SESoM inherits the superior generalization of model ensemble approaches and simultaneously captures the sample-specific competence of each source prompt.
Prompt-Tuning Can Be Much Better Than Fine-Tuning on Cross-lingual Understanding With Multilingual Language Models
Pre-trained multilingual language models show significant performance gains for zero-shot cross-lingual model transfer on a wide range of natural language understanding (NLU) tasks.
Binding Language Models in Symbolic Languages
We propose Binder, a training-free neural-symbolic framework that maps the task input to a program, which (1) allows binding a unified API of language model (LM) functionalities to a programming language (e. g., SQL, Python) to extend its grammar coverage and thus tackle more diverse questions, (2) adopts an LM as both the program parser and the underlying model called by the API during execution, and (3) requires only a few in-context exemplar annotations.
Ranked #4 on
Table-based Fact Verification
on TabFact
FOLIO: Natural Language Reasoning with First-Order Logic
We present FOLIO, a human-annotated, open-domain, and logically complex and diverse dataset for reasoning in natural language (NL), equipped with first order logic (FOL) annotations.
Generating Negative Samples for Sequential Recommendation
Due to the dynamics of users' interests and model updates during training, considering randomly sampled items from a user's non-interacted item set as negatives can be uninformative.
BigIssue: A Realistic Bug Localization Benchmark
As machine learning tools progress, the inevitable question arises: How can machine learning help us write better code?
Policy Optimization for Markov Games: Unified Framework and Faster Convergence
Next, we show that this framework instantiated with the Optimistic Follow-The-Regularized-Leader (OFTRL) algorithm at each state (and smooth value updates) can find an $\mathcal{\widetilde{O}}(T^{-5/6})$ approximate NE in $T$ iterations, and a similar algorithm with slightly modified value update rule achieves a faster $\mathcal{\widetilde{O}}(T^{-1})$ convergence rate.
MACE: An Efficient Model-Agnostic Framework for Counterfactual Explanation
Counterfactual explanation is an important Explainable AI technique to explain machine learning predictions.
Modeling Multi-hop Question Answering as Single Sequence Prediction
Fusion-in-decoder (Fid) (Izacard and Grave, 2020) is a generative question answering (QA) model that leverages passage retrieval with a pre-trained transformer and pushed the state of the art on single-hop QA.
OneAligner: Zero-shot Cross-lingual Transfer with One Rich-Resource Language Pair for Low-Resource Sentence Retrieval
When finetuned on a single rich-resource language pair, be it English-centered or not, our model is able to match the performance of the ones finetuned on all language pairs under the same data budget with less than 2. 0 points decrease in accuracy.
Near-Negative Distinction: Giving a Second Life to Human Evaluation Datasets
Precisely assessing the progress in natural language generation (NLG) tasks is challenging, and human evaluation to establish a preference in a model's output over another is often necessary.
Quiz Design Task: Helping Teachers Create Quizzes with Automated Question Generation
Question generation (QGen) models are often evaluated with standardized NLG metrics that are based on n-gram overlap.
Use All The Labels: A Hierarchical Multi-Label Contrastive Learning Framework
Current contrastive learning frameworks focus on leveraging a single supervisory signal to learn representations, which limits the efficacy on unseen data and downstream tasks.
A Generative Language Model for Few-shot Aspect-Based Sentiment Analysis
Our evaluation results on the single-task polarity prediction show that our approach outperforms the previous state-of-the-art (based on BERT) on average performance by a large margins in few-shot and full-shot settings.
Aspect-Based Sentiment Analysis
Aspect-Based Sentiment Analysis (ABSA)
+4
ELECRec: Training Sequential Recommenders as Discriminators
A generator, as an auxiliary model, is trained jointly with the discriminator to sample plausible alternative next items and will be thrown out after training.
CodeGen: An Open Large Language Model for Code with Multi-Turn Program Synthesis
To democratize this, we train and release a family of large language models up to 16. 1B parameters, called CODEGEN, on natural language and programming language data, and open source the training library JAXFORMER.
Ranked #79 on
Code Generation
on HumanEval
Improving Contrastive Learning with Model Augmentation
However, existing methods all construct views by adopting augmentation from data perspectives, while we argue that 1) optimal data augmentation methods are hard to devise, 2) data augmentation methods destroy sequential correlations, and 3) data augmentation fails to incorporate comprehensive self-supervised signals.
Converse: A Tree-Based Modular Task-Oriented Dialogue System
At the core of the struggle is the need to script every single turn of interactions between the bot and the human user.
ConTinTin: Continual Learning from Task Instructions
This work defines a new learning paradigm ConTinTin (Continual Learning from Task Instructions), in which a system should learn a sequence of new tasks one by one, each task is explained by a piece of textual instruction.
Long Document Summarization with Top-down and Bottom-up Inference
Critical to the success of a summarization model is the faithful inference of latent representations of words or tokens in the source documents.
Ranked #1 on
Text Summarization
on Pubmed
Structure Extraction in Task-Oriented Dialogues with Slot Clustering
Extracting structure information from dialogue data can help us better understand user and system behaviors.
Efficient and Differentiable Conformal Prediction with General Function Classes
Experiments show that our algorithm is able to learn valid prediction sets and improve the efficiency significantly over existing approaches in several applications such as prediction intervals with improved length, minimum-volume prediction sets for multi-output regression, and label prediction sets for image classification.
Intent Contrastive Learning for Sequential Recommendation
Specifically, we introduce a latent variable to represent users' intents and learn the distribution function of the latent variable via clustering.
BLIP: Bootstrapping Language-Image Pre-training for Unified Vision-Language Understanding and Generation
Furthermore, performance improvement has been largely achieved by scaling up the dataset with noisy image-text pairs collected from the web, which is a suboptimal source of supervision.
Ranked #3 on
Open Vocabulary Attribute Detection
on OVAD-Box benchmark
(using extra training data)
UnifiedSKG: Unifying and Multi-Tasking Structured Knowledge Grounding with Text-to-Text Language Models
Structured knowledge grounding (SKG) leverages structured knowledge to complete user requests, such as semantic parsing over databases and question answering over knowledge bases.
Ranked #1 on
Task-Oriented Dialogue Systems
on KVRET
RGRecSys: A Toolkit for Robustness Evaluation of Recommender Systems
Robust machine learning is an increasingly important topic that focuses on developing models resilient to various forms of imperfect data.
QAFactEval: Improved QA-Based Factual Consistency Evaluation for Summarization
Factual consistency is an essential quality of text summarization models in practical settings.
Value Retrieval with Arbitrary Queries for Form-like Documents
Unlike previous methods that only address a fixed set of field items, our method predicts target value for an arbitrary query based on the understanding of the layout and semantics of a form.
Combining Data-driven Supervision with Human-in-the-loop Feedback for Entity Resolution
The distribution gap between training datasets and data encountered in production is well acknowledged.
Open Vocabulary Object Detection with Pseudo Bounding-Box Labels
To enlarge the set of base classes, we propose a method to automatically generate pseudo bounding-box annotations of diverse objects from large-scale image-caption pairs.
Dense Hierarchical Retrieval for Open-Domain Question Answering
In this work, we propose Dense Hierarchical Retrieval (DHR), a hierarchical framework that can generate accurate dense representations of passages by utilizing both macroscopic semantics in the document and microscopic semantics specific to each passage.
Ensemble of Averages: Improving Model Selection and Boosting Performance in Domain Generalization
We first show that this chaotic behavior exists even along the training optimization trajectory of a single model, and propose a simple model averaging protocol that both significantly boosts domain generalization and diminishes the impact of stochasticity by improving the rank correlation between the in-domain validation accuracy and out-domain test accuracy, which is crucial for reliable early stopping.
Ranked #4 on
Domain Generalization
on TerraIncognita
Learning Rich Nearest Neighbor Representations from Self-supervised Ensembles
Pretraining convolutional neural networks via self-supervision, and applying them in transfer learning, is an incredibly fast-growing field that is rapidly and iteratively improving performance across practically all image domains.
Improving Tail-Class Representation with Centroid Contrastive Learning
ICCL interpolates two images from a class-agnostic sampler and a class-aware sampler, and trains the model such that the representation of the interpolative image can be used to retrieve the centroids for both source classes.
Ranked #22 on
Long-tail Learning
on CIFAR-10-LT (ρ=10)
Momentum Contrastive Autoencoder: Using Contrastive Learning for Latent Space Distribution Matching in WAE
Wasserstein autoencoder (WAE) shows that matching two distributions is equivalent to minimizing a simple autoencoder (AE) loss under the constraint that the latent space of this AE matches a pre-specified prior distribution.
MixQG: Neural Question Generation with Mixed Answer Types
Asking good questions is an essential ability for both human and machine intelligence.
DialFact: A Benchmark for Fact-Checking in Dialogue
Fact-checking is an essential tool to mitigate the spread of misinformation and disinformation.
Improving Gender Fairness of Pre-Trained Language Models without Catastrophic Forgetting
In this work, we empirically show that catastrophic forgetting occurs in such methods by evaluating them with general NLP tasks in GLUE.
Robustness Evaluation of Transformer-based Form Field Extractors via Form Attacks
We propose a novel framework to evaluate the robustness of transformer-based form field extraction methods via form attacks.
Field Extraction from Forms with Unlabeled Data
We propose a novel framework to conduct field extraction from forms with unlabeled data.
Self-supervised Learning for Sequential Recommendation with Model Augmentation
However, existing methods all construct views by adopting augmentation from data perspectives, while we argue that 1) optimal data augmentation methods are hard to devise, 2) data augmentation methods destroy sequential correlations, and 3) data augmentation fails to incorporate comprehensive self-supervised signals.
Long Document Summarization with Top-Down and Bottom-Up Representation Inference
Critical to the success of a summarization model is the faithful inference of latent representations of words or tokens in the source documents.
Modeling Dynamic Attributes for Next Basket Recommendation
However, properly integrating them into user interest models is challenging since attribute dynamics can be diverse such as time-interval aware, periodic patterns (etc.
Merlion: A Machine Learning Library for Time Series
We introduce Merlion, an open-source machine learning library for time series.
RnG-KBQA: Generation Augmented Iterative Ranking for Knowledge Base Question Answering
We present RnG-KBQA, a Rank-and-Generate approach for KBQA, which remedies the coverage issue with a generation model while preserving a strong generalization capability.
Contrastive Self-supervised Sequential Recommendation with Robust Augmentation
In this paper, we investigate the application of contrastive Self-Supervised Learning (SSL) to the sequential recommendation, as a way to alleviate some of these issues.
Align before Fuse: Vision and Language Representation Learning with Momentum Distillation
Most existing methods employ a transformer-based multimodal encoder to jointly model visual tokens (region-based image features) and word tokens.
Ranked #5 on
Open Vocabulary Attribute Detection
on OVAD-Box benchmark
(using extra training data)
A Theory-Driven Self-Labeling Refinement Method for Contrastive Representation Learning
Although intuitive, such a native label assignment strategy cannot reveal the underlying semantic similarity between a query and its positives and negatives, and impairs performance, since some negatives are semantically similar to the query or even share the same semantic class as the query.
DocNLI: A Large-scale Dataset for Document-level Natural Language Inference
It has been studied intensively in the past few years thanks to the availability of large-scale labeled datasets.
Learning to Play General-Sum Games Against Multiple Boundedly Rational Agents
We study the problem of training a principal in a multi-agent general-sum game using reinforcement learning (RL).
Understanding the Under-Coverage Bias in Uncertainty Estimation
Estimating the data uncertainty in regression tasks is often done by learning a quantile function or a prediction interval of the true label conditioned on the input.
Policy Finetuning: Bridging Sample-Efficient Offline and Online Reinforcement Learning
This offline result is the first that matches the sample complexity lower bound in this setting, and resolves a recent open question in offline RL.
Are Pretrained Transformers Robust in Intent Classification? A Missing Ingredient in Evaluation of Out-of-Scope Intent Detection
Pre-trained Transformer-based models were reported to be robust in intent classification.
Evaluating State-of-the-Art Classification Models Against Bayes Optimality
Moreover, we show that by varying the temperature of the learned flow models, we can generate synthetic datasets that closely resemble standard benchmark datasets, but with almost any desired Bayes error.
Unsupervised Out-of-Domain Detection via Pre-trained Transformers
Deployed real-world machine learning applications are often subject to uncontrolled and even potentially malicious inputs.
SCRIPT: Self-Critic PreTraining of Transformers
We introduce Self-CRItic Pretraining Transformers (SCRIPT) for representation learning of text.
Controllable Abstractive Dialogue Summarization with Sketch Supervision
In this paper, we aim to improve abstractive dialogue summarization quality and, at the same time, enable granularity control.
BookSum: A Collection of Datasets for Long-form Narrative Summarization
The majority of available text summarization datasets include short-form source documents that lack long-range causal and temporal dependencies, and often contain strong layout and stylistic biases.
QAConv: Question Answering on Informative Conversations
This paper introduces QAConv, a new question answering (QA) dataset that uses conversations as a knowledge source.
Pseudo Siamese Network for Few-shot Intent Generation
PSN consists of two identical subnetworks with the same structure but different weights: an action network and an object network.
Learning to Synthesize Data for Semantic Parsing
Moreover, explicitly modeling compositions using PCFG leads to a better exploration of unseen programs, thus generate more diverse data.
FeTaQA: Free-form Table Question Answering
Existing table question answering datasets contain abundant factual questions that primarily evaluate the query and schema comprehension capability of a system, but they fail to include questions that require complex reasoning and integration of information due to the constraint of the associated short-form answers.
Causal-aware Safe Policy Improvement for Task-oriented dialogue
This method gives guarantees on dialogue policy's performance and also learns to shape rewards according to intentions behind human responses, rather than just mimicking demonstration data; this couple with batch-RL helps overall with sample efficiency of the framework.
Structured Scene Memory for Vision-Language Navigation
Recently, numerous algorithms have been developed to tackle the problem of vision-language navigation (VLN), i. e., entailing an agent to navigate 3D environments through following linguistic instructions.
Sample-Efficient Learning of Stackelberg Equilibria in General-Sum Games
Real world applications such as economics and policy making often involve solving multi-agent games with two unique features: (1) The agents are inherently asymmetric and partitioned into leaders and followers; (2) The agents have different reward functions, thus the game is general-sum.
Local Calibration: Metrics and Recalibration
In this work, we propose the local calibration error (LCE) to span the gap between average and individual reliability.
Don't Just Blame Over-parametrization for Over-confidence: Theoretical Analysis of Calibration in Binary Classification
Modern machine learning models with high accuracy are often miscalibrated -- the predicted top probability does not reflect the actual accuracy, and tends to be over-confident.
Joint Energy-based Model Training for Better Calibrated Natural Language Understanding Models
In this work, we explore joint energy-based model (EBM) training during the finetuning of pretrained text encoders (e. g., Roberta) for natural language understanding (NLU) tasks.
Robustness Gym: Unifying the NLP Evaluation Landscape
Despite impressive performance on standard benchmarks, deep neural networks are often brittle when deployed in real-world systems.
Neural Bayes: A Generic Parameterization Method for Unsupervised Learning
Disjoint Manifold Separation: Neural Bayes allows us to formulate an objective which can optimally label samples from disjoint manifolds present in the support of a continuous distribution.
Noise-Robust Contrastive Learning
In contrast to most existing methods, we combat noise by learning robust representation.
Improved Uncertainty Post-Calibration via Rank Preserving Transforms
In this paper, we propose Neural Rank Preserving Transforms (NRPT), a new post-calibration method that adjusts the output probabilities of a trained classifier using a calibrator of higher capacity, while maintaining its prediction accuracy.
Momentum Contrastive Autoencoder
Quantitatively, we show that our algorithm achieves a new state-of-the-art FID of 54. 36 on CIFAR-10, and performs competitively with existing models on CelebA in terms of FID score.
Learning From Noisy Data With Robust Representation Learning
In contrast to most existing methods, we combat noise by learning robust representation.
ERMAS: Learning Policies Robust to Reality Gaps in Multi-Agent Simulations
Policies for real-world multi-agent problems, such as optimal taxation, can be learned in multi-agent simulations with AI agents that emulate humans.
CorDial: Coarse-to-fine Abstractive Dialogue Summarization with Controllable Granularity
2) A simple strategy to control the granularity of the final summary.
FastIF: Scalable Influence Functions for Efficient Model Interpretation and Debugging
With the availability of the fast influence functions, we demonstrate their usefulness in four applications.
Catastrophic Fisher Explosion: Early Phase Fisher Matrix Impacts Generalization
The early phase of training a deep neural network has a dramatic effect on the local curvature of the loss function.
Bridging Textual and Tabular Data for Cross-Domain Text-to-SQL Semantic Parsing
We present BRIDGE, a powerful sequential architecture for modeling dependencies between natural language questions and relational databases in cross-DB semantic parsing.
Learning from Mistakes: Using Mis-predictions as Harm Alerts in Language Pre-Training
According to recent studies and our empirical observations, one possible reason is that some easy-to-fit patterns in the training data, such as frequently co-occurring word combinations, dominate and harm pre-training, making it hard for the model to fit more complex information.
CTRLsum: Towards Generic Controllable Text Summarization
Our approach enables users to control multiple aspects of generated summaries by interacting with the summarization system through textual input in the form of a set of keywords or descriptive prompts.
GAEA: Graph Augmentation for Equitable Access via Reinforcement Learning
Similarly, social networks within universities and organizations may enable certain groups to more easily access people with valuable information or influence.
Adapt-and-Adjust: Overcoming the Long-Tail Problem of Multilingual Speech Recognition
One crucial challenge of real-world multilingual speech recognition is the long-tailed distribution problem, where some resource-rich languages like English have abundant training data, but a long tail of low-resource languages have varying amounts of limited training data.
CoMatch: Semi-supervised Learning with Contrastive Graph Regularization
CoMatch jointly learns two representations of the training data, their class probabilities and low-dimensional embeddings.
What's New? Summarizing Contributions in Scientific Literature
To overcome this problem, we introduce a new task of disentangled paper summarization, which seeks to generate separate summaries for the paper contributions and the context of the work, making it easier to identify the key findings shared in articles.
Improving Limited Labeled Dialogue State Tracking with Self-Supervision
We present and investigate two self-supervised objectives: preserving latent consistency and modeling conversational behavior.
Probing Task-Oriented Dialogue Representation from Language Models
This paper investigates pre-trained language models to find out which model intrinsically carries the most informative representation for task-oriented dialogue tasks.
Discriminative Nearest Neighbor Few-Shot Intent Detection by Transferring Natural Language Inference
Intent detection is one of the core components of goal-oriented dialog systems, and detecting out-of-scope (OOS) intents is also a practically important skill.
CoCo: Controllable Counterfactuals for Evaluating Dialogue State Trackers
Dialogue state trackers have made significant progress on benchmark datasets, but their generalization capability to novel and realistic scenarios beyond the held-out conversations is less understood.
Ranked #2 on
Multi-domain Dialogue State Tracking
on MULTIWOZ 2.1
(using extra training data)
Unsupervised Paraphrasing with Pretrained Language Models
To enforce a surface form dissimilar from the input, whenever the language model emits a token contained in the source sequence, DB prevents the model from outputting the subsequent source token for the next generation step.
Online Structured Meta-learning
When a new task is encountered, it constructs a meta-knowledge pathway by either utilizing the most relevant knowledge blocks or exploring new blocks.
Explaining and Improving Model Behavior with k Nearest Neighbor Representations
Interpretability techniques in NLP have mainly focused on understanding individual predictions using attention visualization or gradient-based saliency maps over tokens.
How Important is the Train-Validation Split in Meta-Learning?
A common practice in meta-learning is to perform a train-validation split (\emph{train-val method}) where the prior adapts to the task on one split of the data, and the resulting predictor is evaluated on another split.
Towards Theoretically Understanding Why SGD Generalizes Better Than ADAM in Deep Learning
The result shows that (1) the escaping time of both SGD and ADAM~depends on the Radon measure of the basin positively and the heaviness of gradient noise negatively; (2) for the same basin, SGD enjoys smaller escaping time than ADAM, mainly because (a) the geometry adaptation in ADAM~via adaptively scaling each gradient coordinate well diminishes the anisotropic structure in gradient noise and results in larger Radon measure of a basin; (b) the exponential gradient average in ADAM~smooths its gradient and leads to lighter gradient noise tails than SGD.
Representation Learning for Sequence Data with Deep Autoencoding Predictive Components
We propose Deep Autoencoding Predictive Components (DAPC) -- a self-supervised representation learning method for sequence data, based on the intuition that useful representations of sequence data should exhibit a simple structure in the latent space.
Automatic Speech Recognition
Automatic Speech Recognition (ASR)
+3
Universal Natural Language Processing with Limited Annotations: Try Few-shot Textual Entailment as a Start
We demonstrate that this framework enables a pretrained entailment model to work well on new entailment domains in a few-shot setting, and show its effectiveness as a unified solver for several downstream NLP tasks such as question answering and coreference resolution when the end-task annotations are limited.
Discern: Discourse-Aware Entailment Reasoning Network for Conversational Machine Reading
Based on the learned EDU and entailment representations, we either reply to the user our final decision "yes/no/irrelevant" of the initial question, or generate a follow-up question to inquiry more information.
GraPPa: Grammar-Augmented Pre-Training for Table Semantic Parsing
We present GraPPa, an effective pre-training approach for table semantic parsing that learns a compositional inductive bias in the joint representations of textual and tabular data.
Ranked #8 on
Semantic Parsing
on spider
Composed Variational Natural Language Generation for Few-shot Intents
In this paper, we focus on generating training examples for few-shot intents in the realistic imbalanced scenario.
MoPro: Webly Supervised Learning with Momentum Prototypes
We propose momentum prototypes (MoPro), a simple contrastive learning method that achieves online label noise correction, out-of-distribution sample removal, and representation learning.
Ranked #12 on
Image Classification
on OmniBenchmark
(using extra training data)
Photon: A Robust Cross-Domain Text-to-SQL System
Natural language interfaces to databases (NLIDB) democratize end user access to relational data.
SummEval: Re-evaluating Summarization Evaluation
The scarcity of comprehensive up-to-date studies on evaluation metrics for text summarization and the lack of consensus regarding evaluation protocols continue to inhibit progress.
DART: Open-Domain Structured Data Record to Text Generation
Data-to-Text annotations can be a costly process, especially when dealing with tables which are the major source of structured data and contain nontrivial structures.
Explicit Memory Tracker with Coarse-to-Fine Reasoning for Conversational Machine Reading
The goal of conversational machine reading is to answer user questions given a knowledge base text which may require asking clarification questions.
Theory-Inspired Path-Regularized Differential Network Architecture Search
Then we propose a theory-inspired path-regularized DARTS that consists of two key modules: (i) a differential group-structured sparse binary gate introduced for each operation to avoid unfair competition among operations, and (ii) a path-depth-wise regularization used to incite search exploration for deep architectures that often converge slower than shallow ones as shown in our theory and are not well explored during the search.
BERTology Meets Biology: Interpreting Attention in Protein Language Models
Transformer architectures have proven to learn useful representations for protein classification and generation tasks.
Towards Understanding Hierarchical Learning: Benefits of Neural Representations
When the trainable network is the quadratic Taylor model of a wide two-layer network, we show that neural representation can achieve improved sample complexities compared with the raw input: For learning a low-rank degree-$p$ polynomial ($p \geq 4$) in $d$ dimension, neural representation requires only $\tilde{O}(d^{\lceil p/2 \rceil})$ samples, while the best-known sample complexity upper bound for the raw input is $\tilde{O}(d^{p-1})$.
A High-Quality Multilingual Dataset for Structured Documentation Translation
We build and evaluate translation models for seven target languages from English, with several different copy mechanisms and an XML-constrained beam search.
WOAD: Weakly Supervised Online Action Detection in Untrimmed Videos
Online action detection in untrimmed videos aims to identify an action as it happens, which makes it very important for real-time applications.
Ranked #5 on
Online Action Detection
on THUMOS'14
EMT: Explicit Memory Tracker with Coarse-to-Fine Reasoning for Conversational Machine Reading
The goal of conversational machine reading is to answer user questions given a knowledge base text which may require asking clarification questions.
Prototypical Contrastive Learning of Unsupervised Representations
This paper presents Prototypical Contrastive Learning (PCL), an unsupervised representation learning method that addresses the fundamental limitations of instance-wise contrastive learning.
Ranked #5 on
Contrastive Learning
on imagenet-1k
Double-Hard Debias: Tailoring Word Embeddings for Gender Bias Mitigation
Word embeddings derived from human-generated corpora inherit strong gender bias which can be further amplified by downstream models.
ESPRIT: Explaining Solutions to Physical Reasoning Tasks
Our framework learns to generate explanations of how the physical simulation will causally evolve so that an agent or a human can easily reason about a solution using those interpretable descriptions.
VD-BERT: A Unified Vision and Dialog Transformer with BERT
By contrast, in this work, we propose VD-BERT, a simple yet effective framework of unified vision-dialog Transformer that leverages the pretrained BERT language models for Visual Dialog tasks.
TOD-BERT: Pre-trained Natural Language Understanding for Task-Oriented Dialogue
The underlying difference of linguistic patterns between general text and task-oriented dialogue makes existing pre-trained language models less useful in practice.
An investigation of phone-based subword units for end-to-end speech recognition
For Switchboard, our phone-based BPE system achieves 6. 8\%/14. 4\% word error rate (WER) on the Switchboard/CallHome portion of the test set while joint decoding achieves 6. 3\%/13. 3\% WER.
Towards Noise-resistant Object Detection with Noisy Annotations
We address the challenging problem of training object detectors with noisy annotations, where the noise contains a mixture of label noise and bounding box noise.
Differentially Private Deep Learning with Smooth Sensitivity
Ensuring the privacy of sensitive data used to train modern machine learning models is of paramount importance in many areas of practice.
Adv-BERT: BERT is not robust on misspellings! Generating nature adversarial samples on BERT
There is an increasing amount of literature that claims the brittleness of deep neural networks in dealing with adversarial examples that are created maliciously.
Neural Bayes: A Generic Parameterization Method for Unsupervised Representation Learning
Disjoint Manifold Labeling: Neural Bayes allows us to formulate an objective which can optimally label samples from disjoint manifolds present in the support of a continuous distribution.
Explore, Discover and Learn: Unsupervised Discovery of State-Covering Skills
We perform an extensive evaluation of skill discovery methods on controlled environments and show that EDL offers significant advantages, such as overcoming the coverage problem, reducing the dependence of learned skills on the initial state, and allowing the user to define a prior over which behaviors should be learned.
Taylorized Training: Towards Better Approximation of Neural Network Training at Finite Width
We propose \emph{Taylorized training} as an initiative towards better understanding neural network training at finite width.
Proposal Learning for Semi-Supervised Object Detection
two-stage object detectors) by training on both labeled and unlabeled data.
Learning from Noisy Anchors for One-stage Object Detection
State-of-the-art object detectors rely on regressing and classifying an extensive list of possible anchors, which are divided into positive and negative samples based on their intersection-over-union (IoU) with corresponding groundtruth objects.
LiteEval: A Coarse-to-Fine Framework for Resource Efficient Video Recognition
This paper presents LiteEval, a simple yet effective coarse-to-fine framework for resource efficient video recognition, suitable for both online and offline scenarios.
Learning to Retrieve Reasoning Paths over Wikipedia Graph for Question Answering
Answering questions that require multi-hop reasoning at web-scale necessitates retrieving multiple evidence documents, one of which often has little lexical or semantic relationship to the question.
Ranked #26 on
Question Answering
on HotpotQA
Deep Verifier Networks: Verification of Deep Discriminative Models with Deep Generative Models
We test the verifier network on out-of-distribution detection and adversarial example detection problems, as well as anomaly detection problems in structured prediction tasks such as image caption generation.
MKD: a Multi-Task Knowledge Distillation Approach for Pretrained Language Models
Our approach is model agnostic and can be easily applied on different future teacher model architectures.
ERASER: A Benchmark to Evaluate Rationalized NLP Models
We propose several metrics that aim to capture how well the rationales provided by models align with human rationales, and also how faithful these rationales are (i. e., the degree to which provided rationales influenced the corresponding predictions).
Keeping Your Distance: Solving Sparse Reward Tasks Using Self-Balancing Shaped Rewards
For instance, in tasks where the agent must achieve some goal state, simple distance-to-goal reward shaping often fails, as it renders learning vulnerable to local optima.
