6 citations · 26 across the 19 of their papers we have counts for
20 papers
Atomic Skills are the Prerequisite: When Reinforcement Learning Synthesizes Compositional Reasoning, and When It Only Amplifies
Sitao Cheng, Xunjian Yin, Ruiwen Zhou +5
Does Reinforcement Learning (RL) merely amplify existing skills, or synthesize novel skills? We investigate this question through the lens of Complementary Reasoning: the critical…
LEDOM: Reverse Language Model
Xunjian Yin, Sitao Cheng, Yuxi Xie +6
Autoregressive language models are trained exclusively left-to-right. We explore the complementary factorization, training right-to-left at scale, and ask what reasoning patterns e…
How Is LLM Reasoning Distracted by Irrelevant Context? An Analysis Using a Controlled Benchmark
Minglai Yang, Ethan Huang, Liang Zhang +3
We introduce Grade School Math with Distracting Context (GSM-DC), a synthetic benchmark to evaluate Large Language Models' (LLMs) reasoning robustness against systematically contro…
InductionBench: LLMs Fail in the Simplest Complexity Class
Wenyue Hua, Tyler Wong, Sun Fei +3
Large language models (LLMs) have shown remarkable improvements in reasoning and many existing benchmarks have been addressed by models such as o1 and o3 either fully or partially.…
Combating Multimodal LLM Hallucination via Bottom-Up Holistic Reasoning
Shengqiong Wu, Hao Fei, Liangming Pan +3
Recent advancements in multimodal large language models (MLLMs) have shown unprecedented capabilities in advancing various vision-language tasks. However, MLLMs face significant ch…
Aristotle: Mastering Logical Reasoning with A Logic-Complete Decompose-Search-Resolve Framework
Jundong Xu, Hao Fei, Meng Luo +6
In the context of large language models (LLMs), current advanced reasoning methods have made impressive strides in various reasoning tasks. However, when it comes to logical reason…
Understanding the Interplay between Parametric and Contextual Knowledge for Large Language Models
Sitao Cheng, Liangming Pan, Xunjian Yin +2
Large language models (LLMs) encode vast amounts of knowledge during pre-training (parametric knowledge, or PK) and can further be enhanced by incorporating contextual knowledge (C…
Gödel Agent: A Self-Referential Agent Framework for Recursive Self-Improvement
Xunjian Yin, Xinyi Wang, Liangming Pan +3
The rapid advancement of large language models (LLMs) has significantly enhanced the capabilities of AI-driven agents across various tasks. However, existing agentic systems, wheth…
MultiAgent Collaboration Attack: Investigating Adversarial Attacks in Large Language Model Collaborations via Debate
Alfonso Amayuelas, Xianjun Yang, Antonis Antoniades +3
Large Language Models (LLMs) have shown exceptional results on current benchmarks when working individually. The advancement in their capabilities, along with a reduction in parame…
Investigating the Transferability of Code Repair for Low-Resource Programming Languages
Kyle Wong, Alfonso Amayuelas, Liangming Pan +1
Large language models (LLMs) have shown remarkable performance on code generation tasks. A recent use case is iterative code repair, where an LLM fixes an incorrect program by rati…
AKEW: Assessing Knowledge Editing in the Wild
Xiaobao Wu, Liangming Pan, William Yang Wang +1
Knowledge editing injects knowledge updates into language models to keep them correct and up-to-date. However, its current evaluations deviate significantly from practice: their kn…
A Survey on Data Selection for Language Models
Alon Albalak, Yanai Elazar, Sang Michael Xie +11
A major factor in the recent success of large language models is the use of enormous and ever-growing text datasets for unsupervised pre-training. However, naively training a model…
Pride and Prejudice: LLM Amplifies Self-Bias in Self-Refinement
Wenda Xu, Guanglei Zhu, Xuandong Zhao +3
Recent studies show that large language models (LLMs) improve their performance through self-feedback on certain tasks while degrade on others. We discovered that such a contrary i…
Understanding Reasoning Ability of Language Models From the Perspective of Reasoning Paths Aggregation
Xinyi Wang, Alfonso Amayuelas, Kexun Zhang +3
Pre-trained language models (LMs) are able to perform complex reasoning without explicit fine-tuning. To understand how pre-training with a next-token prediction objective contribu…
Position: AI/ML Influencers Have a Place in the Academic Process
Iain Xie Weissburg, Mehir Arora, Xinyi Wang +2
As the number of accepted papers at AI and ML conferences reaches into the thousands, it has become unclear how researchers access and read research publications. In this paper, we…
Efficient Online Data Mixing For Language Model Pre-Training
Alon Albalak, Liangming Pan, Colin Raffel +1
The data used to pretrain large language models has a decisive impact on a model's downstream performance, which has led to a large body of work on data selection methods that aim…
LLMRefine: Pinpointing and Refining Large Language Models via Fine-Grained Actionable Feedback
Wenda Xu, Daniel Deutsch, Mara Finkelstein +6
Recent large language models (LLM) are leveraging human feedback to improve their generation quality. However, human feedback is costly to obtain, especially during inference. In t…
MAF: Multi-Aspect Feedback for Improving Reasoning in Large Language Models
Deepak Nathani, David Wang, Liangming Pan +1
Language Models (LMs) have shown impressive performance in various natural language tasks. However, when it comes to natural language reasoning, LMs still face challenges such as h…
INSTRUCTSCORE: Explainable Text Generation Evaluation with Finegrained Feedback
Wenda Xu, Danqing Wang, Liangming Pan +4
Automatically evaluating the quality of language generation is critical. Although recent learned metrics show high correlation with human judgement, these metrics can not explain t…
The Knowledge Alignment Problem: Bridging Human and External Knowledge for Large Language Models
Shuo Zhang, Liangming Pan, Junzhou Zhao +1
Large language models often necessitate grounding on external knowledge to generate faithful and reliable answers. Yet even with the correct groundings in the reference, they can i…