Articles from HuggingFace
HypoEvolve introduces a framework using genetic algorithms to coordinate multi-agent LLMs for scientific hypothesis discovery, demonstrated through drug repurposing in cancer research with improved performance over baselines.
HarnessVLN is a zero-shot, training-free framework for embodied navigation that unifies perception, retrieval, grounding, navigation, recovery, and termination through a unified tool interface, achieving state-of-the-art results on benchmarks like R2R and RxR.
ModularRSI introduces a modular and generalizable framework for recursive self-improvement in AI agent harnesses, using contrastive learning across tasks to evolve modules independently and enhance performance on unseen tasks.
The paper introduces Gavel, a method that elicits native skill routing from frozen LLMs via linear projections, enabling efficient tool selection without context overload and outperforming existing pipelines on benchmarks.
This paper decomposes transformer representation updates into parallel and perpendicular components to study evolution geometry, linking it to editing robustness, compression diagnosis, and training improvements.
ModaLens is a paired image-swap audit that measures how report availability reduces image sensitivity in medical vision-language models, demonstrated using MedGemma-27B on the MIMIC-CXR dataset.
This paper investigates the losslessness of Orthrus, a hybrid autoregressive-diffusion model for inference acceleration, finding that it requires high numerical precision (FP32) for exact trajectory matching, while BF16 divergence does not impair downstream performance.
LynnReal-Omni is a unified multimodal video diffusion framework that integrates agentic visual controls with high-fidelity generation and real-time acceleration for stable, controllable video creation.
This paper proposes an exploration-guided prompt scaffolding framework that dynamically adjusts training prompts for multimodal reinforcement learning, achieving up to 9.7% relative improvement in performance on benchmarks.
PhysBrain 1.5 is a unified model that integrates physical environment understanding, action generation, and future state prediction via autoregressive training, achieving state-of-the-art open-source performance on 28 embodied benchmarks.
KaiNinja extends a native 3D generator to produce part-level outputs using a dual-volume representation, improving both part and whole-object fidelity without requiring segmentation.
Atria Dawn Preview is a foundation agentic language model designed for scientific research, achieving competitive benchmark results and demonstrating a shift toward human-AI project-level collaboration.
This paper introduces Elo-per-token analysis to study how LLM agents allocate test-time compute, revealing that agents initially outperform independent sampling but slow down over time, with parallel sessions offering performance gains.
This paper introduces a method for vibe design agents to explore diverse UI alternatives by separating exploration from implementation through structured design specifications, as evaluated on 168 prompts and a large online experiment with over 300,000 tasks.
This paper introduces SAILS, a method for selecting optimal poison sets in backdoor attacks against large language models, improving worst-case attack success by 30 percentage points over baselines.
HazardAuditor introduces an execution-grounded framework for supervising safety in computer-use agents, with Guard Policy Optimization improving safety outcomes by up to 16.5% accuracy over prior methods.
RSIAgent is a training-free multi-agent framework that enables digital agents to adapt to new environments through recursive self-improvement, autonomous memory construction, and broad-then-deep exploration, outperforming closed-source models on benchmarks.
Omni-Streaming Thinking improves streaming omni-modal reasoning by deferring claims until cross-modal verification, reducing premature commitment and auditory hallucinations.
Dream-RSI is a framework for scalable recursive self-improvement in AI agents that uses historical discovery trees to create a replay simulator for offline policy evaluation, reducing online costs and improving discovery efficiency.
The paper introduces BVB, a benchmark for agentic video understanding via programmatic reconstruction in Blender, evaluating models on perceptual similarity and spatiotemporal fact retention.