This work asks why evidence integration in large language models requires thinking, and reasoning trajectories showed models revisiting input, recounting letters, and checking intermediate counts that informed the answer, suggesting that thinking constructs the accumulated count that direct responses lack rather than r...
Jing-Ming Xue, Robert C. Wilson, Hua-Dong Xiong· 0 citations
In this controlled decision setting, thinking improved how models acted on current evidence, while neither measured signature supported a shift toward a more information-seeking policy.
Hua-Dong Xiong, Xinyuan Yan, Ji-An Li et al.· arXiv.org· 0 citations
It is proved that the population gap is zero when the partner's choices convey nothing beyond their literal content, given a fair shuffle and a player who uses only what was said and seen, and argued for reporting the convention gap on human-partnered play, not only AI-AI performance, when evaluating agents intended fo...
Makoto Fukushima, Hua-Dong Xiong, E. M. Pari· 1 citation· ⚡1
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.