HuggingFace Daily Papers — 2026-06-01

Top upvoted daily papers on HuggingFace.

SoundnessBench: Can Your AI Scientist Really Tell Good Research Ideas from Bad Ones?

Upvotes: 0 | Published: 2026-05-27

Emergent Languages in Populations of Language Model Agents: From Token Efficiency to Oversight Evasion

Upvotes: 0 | Published: 2026-05-28

PEEK: Picking Essential frames via Efficient Knowledge distillation

Upvotes: 0 | Published: 2026-05-28

iVGR: Internalizing Visually Grounded Reasoning for MLLMs with Reinforcement Learning

Upvotes: 0 | Published: 2026-05-28

Benchmarking Composed Image Retrieval for Applied Earth Observation

Upvotes: 0 | Published: 2026-05-22

DRIFT: Decoupled Rollouts and Importance-Weighted Fine-Tuning for Efficient Multi-Turn Optimization

Upvotes: 0 | Published: 2026-05-28

Recovering Policy-Induced Errors: Benchmarking and Trajectory Synthesis for Robust GUI Agents

Upvotes: 0 | Published: 2026-05-27

Seeing Isn't Knowing: Do VLMs Know When Not to Answer Spatial Questions (and Why)?

Upvotes: 0 | Published: 2026-05-27

SCOPE: Self-Play via Co-Evolving Policies for Open-Ended Tasks

Upvotes: 0 | Published: 2026-05-28

The Flip Side of RLHF: On-Policy Feedback for Reward Model Self-Supervised Improvement

Upvotes: 0 | Published: 2026-05-28

Count Anything

Upvotes: 0 | Published: 2026-05-28

AnyMo: Scaling Any-Modality Conditional Motion Generation with Masked Modeling

Upvotes: 0 | Published: 2026-05-27

The Good, the Bad, and the Ugly of Markov Boundary for Tabular Prediction

Upvotes: 0 | Published: 2026-05-27

Frequency-Guided Action Diffusion via Sub-Frequency Manifold Traversal

Upvotes: 0 | Published: 2026-05-26

Comprehensive Benchmarking of Long-Form Speech Generation in Diverse Scenarios

Upvotes: 0 | Published: 2026-05-26

Towards Streaming Synchronized Spatial Audio Generation via Autoregressive Diffusion Transformer

Upvotes: 0 | Published: 2026-05-28

SwanVoice: Expressive Long-Form Zero-Shot Speech Synthesis for Both Monologue and Dialogue

Upvotes: 0 | Published: 2026-05-28

FRAPPE: Full Input, Residual Output Autoencoding with Projection Pursuit Encoder

Upvotes: 0 | Published: 2026-05-26

OpenSkillEval: Automatically Auditing the Open Skill Ecosystem for LLM Agents

Upvotes: 0 | Published: 2026-05-27

COLLEAGUE.SKILL: Automated AI Skill Generation via Expert Knowledge Distillation

Upvotes: 0 | Published: 2026-05-28