arXiv AI By Yuming (Rapheal), Huang, Yao Liu, Pengjie Ding, Lei Wang, Junchen Wan

Does Capability Transfer to Subjective Behavior -- and Would Our Instruments Tell Us? A Self-Evolving, Trust-by-Construction Evaluation Paradigm

Read the original on arXiv AI →

arXiv:2605. 27914v2 Announce Type: replace-cross Abstract: Benchmarking is mature where answers are verifiable -- math, code, reasoning -- but the fastest-growing uses of LLMs are subjective and human-facing: companionship, emotional support, counseling.

Summary generated by The Flow from the publisher's feed. The full article lives at arXiv AI.

arXiv AI
2d ago

Inducing Reward-Free Judging Rubrics that Reduce Over-Crediting in Agent Evaluation

arXiv:2608. 13564v1 Announce Type: new Abstract: Evaluating language-model agents at scale increasingly relies on a second language model as an automatic judge, because the gold signal, an executable environment reward, is expensive, slow, or unavailable at deployment time.

By Darragh Quinn, David Dylan, Roisin Healy, Fionn Carroll, Maeve Donnelly, Cormac Sheehan