Zobrazeno 1 - 10
of 24
pro vyhledávání: '"Zhang, Hugh"'
Autor:
Wang, Evan, Cassano, Federico, Wu, Catherine, Bai, Yunfeng, Song, Will, Nath, Vaskar, Han, Ziwen, Hendryx, Sean, Yue, Summer, Zhang, Hugh
While scaling training compute has led to remarkable improvements in large language models (LLMs), scaling inference compute has not yet yielded analogous gains. We hypothesize that a core missing component is a lack of diverse LLM outputs, leading t
Externí odkaz:
http://arxiv.org/abs/2409.03733
Autor:
Li, Nathaniel, Han, Ziwen, Steneker, Ian, Primack, Willow, Goodside, Riley, Zhang, Hugh, Wang, Zifan, Menghini, Cristina, Yue, Summer
Recent large language model (LLM) defenses have greatly improved models' ability to refuse harmful queries, even when adversarially attacked. However, LLM defenses are primarily evaluated against automated adversarial attacks in a single turn of conv
Externí odkaz:
http://arxiv.org/abs/2408.15221
Autor:
Nath, Vaskar, Slack, Dylan, Da, Jeff, Ma, Yuntao, Zhang, Hugh, Whitehead, Spencer, Hendryx, Sean
Techniques that learn improved representations via offline data or self-supervised objectives have shown impressive results in traditional reinforcement learning (RL). Nevertheless, it is unclear how improved representation learning can benefit reinf
Externí odkaz:
http://arxiv.org/abs/2407.13887
Autor:
Zheng, Huaixiu Steven, Mishra, Swaroop, Zhang, Hugh, Chen, Xinyun, Chen, Minmin, Nova, Azade, Hou, Le, Cheng, Heng-Tze, Le, Quoc V., Chi, Ed H., Zhou, Denny
We introduce NATURAL PLAN, a realistic planning benchmark in natural language containing 3 key tasks: Trip Planning, Meeting Planning, and Calendar Scheduling. We focus our evaluation on the planning capabilities of LLMs with full information on the
Externí odkaz:
http://arxiv.org/abs/2406.04520
Autor:
Zhang, Hugh, Da, Jeff, Lee, Dean, Robinson, Vaughn, Wu, Catherine, Song, Will, Zhao, Tiffany, Raja, Pranav, Slack, Dylan, Lyu, Qin, Hendryx, Sean, Kaplan, Russell, Lunati, Michele, Yue, Summer
Large language models (LLMs) have achieved impressive success on many benchmarks for mathematical reasoning. However, there is growing concern that some of this performance actually reflects dataset contamination, where data closely resembling benchm
Externí odkaz:
http://arxiv.org/abs/2405.00332
Autor:
Li, Kenneth, Jelassi, Samy, Zhang, Hugh, Kakade, Sham, Wattenberg, Martin, Brandfonbrener, David
We present an approach called Q-probing to adapt a pre-trained language model to maximize a task-specific reward function. At a high level, Q-probing sits between heavier approaches such as finetuning and lighter approaches such as few shot prompting
Externí odkaz:
http://arxiv.org/abs/2402.14688
We propose ABCs (Adaptive Branching through Child stationarity), a best-of-both-worlds algorithm combining Boltzmann Q-learning (BQL), a classic reinforcement learning algorithm for single-agent domains, and counterfactual regret minimization (CFR),
Externí odkaz:
http://arxiv.org/abs/2402.11835
Autor:
Zhang, Hugh, Parkes, David C.
Large language models have astounded the world with fascinating new capabilities. However, they currently lack the ability to teach themselves new skills, relying instead on large amounts of human-generated training data. We introduce SECToR (Self-Ed
Externí odkaz:
http://arxiv.org/abs/2309.08589
Autor:
Zhang, Hugh
Simple adaptive procedures that converge to correlated equilibria are known to exist for normal form games (Hart and Mas-Colell 2000), but no such analogue exists for extensive-form games. Leveraging inspiration from Zinkevich et al. (2008), we show
Externí odkaz:
http://arxiv.org/abs/2207.06548
We extend the classic regret minimization framework for approximating equilibria in normal-form games by greedily weighing iterates based on regrets observed at runtime. Theoretically, our method retains all previous convergence rate guarantees. Empi
Externí odkaz:
http://arxiv.org/abs/2204.04826