1 citations · 1 across the 8 of their papers we have counts for
7 papers · 1 filter
Mitigating Exploration Bias in RL for Multi-Instruction Following
Mian Zhang, Yueqin Yin, Kaiyu He +4
RL has emerged as a powerful paradigm for enhancing the instruction following capabilities of LLMs. While existing training recipes achieve substantial gains, we find that they suf…
Is Grokking Worthwhile? Functional Analysis and Transferability of Generalization Circuits in Transformers
Kaiyu He, Zhang Mian, Peilin Wu +2
While Large Language Models (LLMs) excel at factual retrieval, they often struggle with the "curse of two-hop reasoning" in compositional tasks. Recent research suggests that param…
HiPRAG: Hierarchical Process Rewards for Efficient Agentic Retrieval Augmented Generation
Peilin Wu, Mian Zhang, Kun Wan +4
Agentic RAG is a powerful technique for incorporating external information that LLMs lack, enabling better problem solving and question answering. However, suboptimal search behavi…
GEAR: A General Evaluation Framework for Abductive Reasoning
Kaiyu He, Peilin Wu, Mian Zhang +4
Since the advent of large language models (LLMs), research has focused on instruction following and deductive reasoning. A central question remains: can these models discover new k…
Semantic Pivots Enable Cross-Lingual Transfer in Large Language Models
Kaiyu He, Tong Zhou, Yubo Chen +4
Large language models (LLMs) demonstrate remarkable ability in cross-lingual tasks. Understanding how LLMs acquire this ability is crucial for their interpretability. To quantify t…
From Reasoning to Learning: A Survey on Hypothesis Discovery and Rule Learning with Large Language Models
Kaiyu He, Zhiyu Chen
Since the advent of Large Language Models (LLMs), efforts have largely focused on improving their instruction-following and deductive reasoning abilities, leaving open the question…
IDEA: Enhancing the Rule Learning Ability of Large Language Model Agent through Induction, Deduction, and Abduction
Kaiyu He, Mian Zhang, Shuo Yan +2
While large language models (LLMs) have been thoroughly evaluated for deductive and inductive reasoning, their proficiency in holistic rule learning in interactive environments rem…