Cooperative language-model agents can improve without weight updates by revising reusable skills, but only if a shared team reward is traced to the responsible agent and skill. Existing methods credit the skills present in each trajectory, yet presence is not responsibility: a harmful skill active in successful runs st...
Ya-Nan Ma, Yang Wang, David Le et al.· 0 citations
JudgeMoE, a lightweight aggregator that assigns example-specific weights to cached judge score distributions and fuses them before computing a final score is introduced, showing that score-range choice is unstable across judge--dataset settings and that soft scoring usually outperforms hard decoding.
Yiqi Liu, Joseph James, Yang Wang et al.· 0 citations
When a language model refuses to answer a prompt, it is unclear whether the correct answer is erased from its internal representations, or merely suppressed at the output layer. We investigate this mechanism using a controlled withhold setting, which yields perfectly matched answering and refusal trajectories for bidir...
Yiqi Liu, Yang Wang, Songxin Wang et al.· 0 citations
This work proposes ToolScout, a source-aware routing method that uses query-side TF-IDF fingerprints flag source styles on which the fine-tuned retriever is likely to fail better than semantic or length-based proxies, giving a cheap signal for mismatch over a fixed tool corpus.
Yiqi Liu, Joseph James, Yang Wang et al.· 2 citations
A benchmark for evaluating whether models can infer the implicit, non-linear, and rhetorically layered meanings of social media videos that appear nonsensical on the surface but convey deliberate pragmatic meanings, and a diagnostic setting for measuring the gap between multimodal perception and pragmatic comprehension...
Yang Wang, Ya-Nan Ma, Yiqi Liu et al.· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
