We have 7 of 91 papers
We haven’t gathered this author’s papers yet. Follow them and we’ll fetch their work.
Not the right person? Other researchers publish under this name.
Building reliable robot capabilities across diverse tasks requires substantial human effort to develop and maintain skills, design rewards, and integrate perception with control. We present Reconstruct, Practice, Go Real (RPG), a framework for autonomous improvement of robot execution systems without updating model wei...
When inference demand exceeds available compute capacity, model providers must decide which requests should be served first. Users have different tolerances for delay from an LLM API, but current priority pricing schemes compress these differences into coarse fixed-price service tiers. We design an inference auction th...
The results strictly weaken the assumptions required by prior work in the multi-agent information aggregation literature, filling a gap that had remained elusive even for games with constant $CC_\alpha(G)$.
This work proposes Abstract Token Curriculum (ATC), a novel curriculum learning framework that elicits effective continuous intermediate representations without direct supervision or manual scratchpad design and shows ATC's effectiveness on graph reachability and arithmetic learning tasks.
This work presents a fine-grained analysis of the distortion of RLHF with reward clipping and demonstrates that such exponential degradation is not a fundamental property of the algorithm but rather a consequence of distribution mismatch between the distribution generating preference data and the KL reference policy.
The notion of assistance regret is introduced: the gap between the cumulative utility of interactions and that of the optimal joint policies in hindsight, which map latent states to action pairs, is introduced.
This work proposes a game-theoretic framework that gives this reward-retention trade-off an explicit statistical interpretation, and provides a principled method for learning this equilibrium coefficient via reduction to the KL-regularized RL objective, thus allowing for flexible integration into standard fine-tuning p...
We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.
