4 papers
When Harnesses Lose the Signal: Causal Evaluation of Recovery in LLM Agents
Shuyao Xiao, Shengling Wang, Xuan Chen +7
Large language model agents rely on external harnesses to pass information between the model and its environment and to recover from execution errors. Yet recovery is usually judge…
Search Shapes Conclusions: Auditing Evidence Selection Bias in Deep Research Agents
Shuyao Xiao, Shengling Wang, Xuan Chen +8
Deep Research agents synthesize evidence into cited reports, yet a well-cited report can still reach a misleading conclusion. Citation correctness checks whether cited sources supp…
Certifying Model Upgrades with Slice-Wise Non-Regression and Incumbent Fallback
Shengwei Zhang, Tao Wu, Fei Qian
An updated model can improve an aggregate metric while degrading a slice that matters to a downstream user. We study checkpoint selection subject to non-regression tolerances relat…
When Errors Become Memories: Causal Pathway Tracing in Multi-Turn Memory-Augmented LLMs
Shuyao Xiao, Shengling Wang, Xuan Chen +8
Long-term memory enables large language models (LLMs) to preserve and reuse information across interactions, but it can also turn localized errors into persistent risks. Existing w…