14 papers
TRAJDEBUG: Tracing Error Lifecycle to Identify Critical Failures in Long-Horizon Agent Trajectories
Yunjia Qi, Zehua Yin, Xintong Shi +10
LLM-based agentic systems have shown remarkable capabilities in complex domains, while suffering from cascading errors and difficulty in debugging. Critical error detection aims to…
Can LLM-as-a-Judge Reliably Verify Rubrics in Agentic Scenarios?
Yangda Peng, Yunjia Qi, Hao Peng +11
Rubric-based scoring has become a widely used paradigm in model evaluation, typically with LLM-as-a-Judge (LaaJ) for rubric scoring. However, the reliability of LaaJ for rubric sco…
LongTraceRL: Learning Long-Context Reasoning from Search Agent Trajectories with Rubric Rewards
Nianyi Lin, Jiajie Zhang, Lei Hou +1
Long-context reasoning remains a central challenge for large language models, which often fail to locate and integrate key information in extensive distracting content. Reinforceme…
Guiding LLM Post-training Data Engineering with Model Internals from Sparse Autoencoders
Yi Jing, Zao Dai, Jinwu Hu +4
Model internals encode rich information about how a large language model (LLM) processes its training data; however, post-training data engineering largely relies on external signa…
StoryAlign: Evaluating and Training Reward Models for Story Generation
Haotian Xia, Hao Peng, Yunjia Qi +4
Story generation aims to automatically produce coherent, structured, and engaging narratives. Although large language models (LLMs) have significantly advanced text generation, sto…
WildReward: Learning Reward Models from In-the-Wild Human Interactions
Hao Peng, Yunjia Qi, Xiaozhi Wang +3
Reward models (RMs) are crucial for the training of large language models (LLMs), yet they typically rely on large-scale human-annotated preference pairs. With the widespread deplo…