…learning and dense supervision, using sparse rewards for teacher model discovery and dense rewards for student model compression. AI-generated summary In settings where labeled verifiable training data is the binding constraint…
…Grounded in large-scale studies of real coding-agent interactions, this setup tests whether agents can discover user intent, adapt to evolving requirements, and build on their own prior work. Across a…
…We ask whether frontier-level reasoning performance can be improved without additional training, by reorganizing latent capabilities already encoded in existing checkpoints. Darwin introduces three key ideas: (i) a 14-dimensional adaptive…
…Extensive experiments demonstrate that TriGlue generates chemically valid molecules and produces plausible ternary complexes, which highlight the potential of biology-inspired generative modeling for accelerating molecular glue discovery. Our code is available…
…and reward replicating known optimizations rather than discovering new ones. The resulting reward signals are misleading: agents learn to generate kernels that score well in sandboxes but introduce interface incompatibilities, compilation-stack…
…task in parallel, the hub records concise notes on what each agent has established, attempted, or ruled out, and enables each agent to selectively access what other agents have discovered in a…
…AI-generated summary This report describes ARIS (Auto-Research-in-sleep), an open-source research harness for autonomous research, including its architecture, assurance mechanisms, and early deployment experience. The performance of agent…
…Jun 11 Effective, Efficient, and Explainable AI Lab Authors: , Abstract Independent component analysis (ICA) is revived as an efficient method for discovering interpretable directions in language model representations, offering a faster alternative…
…human scientific discovery. It’s also relevant given the rapid progress in generative AI, with models like Sora setting benchmarks, and Hugging Face’s role as a leader in making these tools…
…Concretely, we propose masked boundary modeling , a self-supervised paradigm that dynamically learns sub-pixel boundary representations and subsequently leverages the discovered boundary-bearing tokens as masked targets to facilitate dense visual…