← Back to all news
arXiv AI By Weilin Jin, Mingyu Wang, Taiyu Zhu, Ziqi Zhou, Wenbo Li, Haoyang Huang, Nan Duan, Yifan Wu, Ying Li, Zhonghai Wu

ReCast: Attribution-Oriented Step Representation Learning for LLM-Based Agent Systems

Read the original on arXiv AI →

ReCast is a new method for learning step representations in large language model (LLM) based agent systems, aimed at improving failure attribution by transforming hidden states into attribution‑oriented representations. The approach selects relevant layers, builds pattern and deviation features, and trains an encoder with contrastive and ranking objectives. In experiments on four benchmarks, ReCast outperforms the strongest baseline, achieving higher Hit@1 scores on the Who&When Algorithm and Handcrafted datasets.

Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv AI.

Related stories

arXiv Machine Learning
Oct 12

Can Failed Trajectories Reveal Where Agents Go Wrong? Agentic Failure Attribution Without Step-Level Labels

The paper introduces SOAP, a method for attributing failures in large language model agents without requiring step-level labels or a separate set of successful trajectories. SOAP builds a spectral reference space from LLM representations to assign base error scores to each step, then uses attention-based soft dependencies to propagate downstream error signals back to their likely upstream causes. Experiments demonstrate that SOAP outperforms existing baselines in accurately identifying decisive steps responsible for task failures.

By Do Nguyen-Thanh, Junlin Fang, Bing Li, Zhen Fang, Sean Du
llmsagents
More like this →
arXiv AI
Sep 7

DCFA: Dual-view Causal-inspired Attribution for Failure Reasoning in LLM-based Multi-agent Systems

The paper introduces DCFA, a training‑free framework for attributing failures in large language model‑based multi‑agent systems. DCFA uses a global module to build causal‑inspired dependency graphs from system traces, pinpointing the earliest decisive error, and a local module that refines this attribution through counterfactual reasoning. Experiments on the Who&When benchmark across six LLMs demonstrate that DCFA improves step‑level accuracy by up to 8.27% over existing baselines.

By Zehao Wang, Lanjun Wang, Shilong Jin, Junjie Chen, Yanghua Xiao
llmsagentsbenchmarks
More like this →