23 citations · 99 across the 26 of their papers we have counts for
20 papers · 1 filter
RAG-Zeval: Towards Robust and Interpretable Evaluation on RAG Responses through End-to-End Rule-Guided Reasoning
Kun Li, Yunxiang Li, Tianhua Zhang +4
Robust evaluation is critical for deploying trustworthy retrieval-augmented generation (RAG) systems. However, current LLM-based evaluation frameworks predominantly rely on directl…
Generate, Discriminate, Evolve: Enhancing Context Faithfulness via Fine-Grained Sentence-Level Self-Evolution
Kun Li, Tianhua Zhang, Yunxiang Li +5
Improving context faithfulness in large language models is essential for developing trustworthy retrieval augmented generation systems and mitigating hallucinations, especially in…
Decoding on Graphs: Faithful and Sound Reasoning on Knowledge Graphs through Generation of Well-Formed Chains
Kun Li, Tianhua Zhang, Xixin Wu +3
Knowledge Graphs (KGs) can serve as reliable knowledge sources for question answering (QA) due to their structured representation of knowledge. Existing research on the utilization…
Adaptive Query Rewriting: Aligning Rewriters through Marginal Probability of Conversational Answers
Tianhua Zhang, Kun Li, Hongyin Luo +3
Query rewriting is a crucial technique for passage retrieval in open-domain conversational question answering (CQA). It decontexualizes conversational queries into self-contained q…
Self-MoE: Towards Compositional Large Language Models with Self-Specialized Experts
Junmo Kang, Leonid Karlinsky, Hongyin Luo +7
We present Self-MoE, an approach that transforms a monolithic LLM into a compositional, modular system of self-specialized experts, named MiXSE (MiXture of Self-specialized Experts…
THREAD: Thinking Deeper with Recursive Spawning
Philip Schroeder, Nathaniel Morgan, Hongyin Luo +1
Large language models (LLMs) have shown impressive capabilities across diverse settings, but still struggle as the length and complexity of the context increases. To address this c…
Self-Specialization: Uncovering Latent Expertise within Large Language Models
Junmo Kang, Hongyin Luo, Yada Zhu +6
Recent works have demonstrated the effectiveness of self-alignment in which a large language model is aligned to follow general instructions using instructional data generated from…
Natural Language Embedded Programs for Hybrid Language Symbolic Reasoning
Tianhua Zhang, Jiaxin Ge, Hongyin Luo +7
How can we perform computations over natural language representations to solve tasks that require symbolic and numeric reasoning? We propose natural language embedded programs (NLE…
DoLa: Decoding by Contrasting Layers Improves Factuality in Large Language Models
Yung-Sung Chuang, Yujia Xie, Hongyin Luo +3
Despite their impressive capabilities, large language models (LLMs) are prone to hallucinations, i.e., generating content that deviates from facts seen during pretraining. We propo…
SAIL: Search-Augmented Instruction Learning
Hongyin Luo, Yung-Sung Chuang, Yuan Gong +6
Large language models (LLMs) have been significantly improved by instruction fine-tuning, but still lack transparency and the ability to utilize up-to-date knowledge and informatio…
Entailment as Robust Self-Learner
Jiaxin Ge, Hongyin Luo, Yoon Kim +1
Entailment has been recognized as an important metric for evaluating natural language understanding (NLU) models, and recent studies have found that entailment pretraining benefits…
Interpretable Unified Language Checking
Tianhua Zhang, Hongyin Luo, Yung-Sung Chuang +7
Despite recent concerns about undesirable behaviors generated by large language models (LLMs), including non-factual, biased, and hateful language, we find LLMs are inherent multi-…
Logic Against Bias: Textual Entailment Mitigates Stereotypical Sentence Reasoning
Hongyin Luo, James Glass
Due to their similarity-based learning objectives, pretrained sentence encoders often internalize stereotypical assumptions that reflect the social biases that exist within their t…
DiffCSE: Difference-based Contrastive Learning for Sentence Embeddings
Yung-Sung Chuang, Rumen Dangovski, Hongyin Luo +7
We propose DiffCSE, an unsupervised contrastive learning framework for learning sentence embeddings. DiffCSE learns sentence embeddings that are sensitive to the difference between…
Mitigating Biases in Toxic Language Detection through Invariant Rationalization
Yung-Sung Chuang, Mingye Gao, Hongyin Luo +4
Automatic detection of toxic language plays an essential role in protecting social media users, especially minority groups, from verbal abuse. However, biases toward some attribute…
Meta-learning for downstream aware and agnostic pretraining
Hongyin Luo, Shuyan Dong, Yung-Sung Chuang +1
Neural network pretraining is gaining attention due to its outstanding performance in natural language processing applications. However, pretraining usually leverages predefined ta…
Cooperative Self-training of Machine Reading Comprehension
Hongyin Luo, Shang-Wen Li, Mingye Gao +2
Pretrained language models have significantly improved the performance of downstream language understanding tasks, including extractive question answering, by providing high-qualit…
Knowledge Grounded Conversational Symptom Detection with Graph Memory Networks
Hongyin Luo, Shang-Wen Li, James Glass
In this work, we propose a novel goal-oriented dialog task, automatic symptom detection. We build a system that can interact with patients through dialog to detect and collect clin…
Prototypical Q Networks for Automatic Conversational Diagnosis and Few-Shot New Disease Adaption
Hongyin Luo, Shang-Wen Li, James Glass
Spoken dialog systems have seen applications in many domains, including medical for automatic conversational diagnosis. State-of-the-art dialog managers are usually driven by deep…
Improving Neural Language Models by Segmenting, Attending, and Predicting the Future
Hongyin Luo, Lan Jiang, Yonatan Belinkov +1
Common language models typically predict the next word given the context. In this work, we propose a method that improves language modeling by learning to align the given context a…