2 papers
cs.LG2026
Multimodal Evaluator Preference Collapse: Cross-Modal Coupling in Self-Evolving Agents
Zewen Liu
When AI agents use language models to evaluate their own outputs in a feedback loop, systematic biases emerge. We show that Evaluator Preference Collapse (EPC) is dramatically ampl…
cs.AI2026
Harness Updating Is Not Harness Benefit: Disentangling Evolution Capabilities in Self-Evolving LLM Agents
Minhua Lin, Juncheng Wu, Zijun Wang +14
LLM agents are increasingly deployed as systems built around editable external harnesses, including prompts, skills, memories and tools, that shape task execution without changing…