Research Papers
OneStreamer: A 4B Model That Learns What to Remember and When to Respond in Streaming Video
Hugging Face researchers introduce OneStreamer, a 4B streaming video LLM that jointly learns query-independent evidence recording...
New Study Finds Post-Training 'Sharpening Tax' Limits LLM Test-Time Scaling
Researchers propose a diagnostic metric called 'Sharpening Tax' showing that RL post-training improves single-shot accuracy but re...
World Observer Decouples Watching from Acting to Keep World Models Consistent
A KAIST CVLab paper introduces World Observer, a framework that pairs an actor-centric view with panoramic observers so objects re...
Hugging Face Paper Proposes 'Rhetorical Robustness' as New Test for AI Peer Reviewers
A new paper introduces Rhetorical Robustness as a joint requirement of stability under content-preserving rewrites and discriminat...
APPL Uses Structural Priors to Bridge Skill Learning and Composition in Robots
A new Hugging Face paper introduces Agent Priors-guided Policy Learning (APPL), which treats each policy's structural prior as par...
Hugging Face Paper Introduces HC-DLM, a Hybrid Diffusion Language Model
A new research paper proposes Hierarchical Continuous Diffusion Language Models (HC-DLM), which combine discrete token generation...
Hugging Face Paper Proposes Adaptive Reward Routing for Joint Audio-Video Diffusion
A new research paper introduces Adaptive Reward Routing, a method that dynamically adjusts where reward-driven updates act and how...
LANTERN Uses Model Activations to Uncover Hidden Math Links in OEIS
A new pipeline called LANTERN ranks candidate relations between integer sequences using a classifier over pretrained-model activat...
Hugging Face Paper Unveils 50,000-Pair Agent Error Dataset for Failure-Aware Training
The Agent Error Dataset (AED) compiles 50,228 error-diagnosis pairs drawn from 9,961 tasks across 33 environments, and its five-st...
Hugging Face Paper Introduces AIM: An Agentic Framework for Managing Research Ideas
A new paper proposes the Agentic Idea Manager (AIM), a fully autonomous framework that organizes, selects, and audits research ide...
LoopVL Brings Recurrent Computation to Vision-Language Models
Hugging Face paper LoopVL explores whether Loop Transformers can be extended to vision-language models, combining Module-Loop and...
Hugging Face Paper Exposes 'Ordinal Scale-Utilization Bias' in Direct-Decision Models
A new study of JEV 1.13 and three open KEV models finds that direct-decision models compress the ordinal scales users supply, usin...