Crimson AI NewsA CrimsonLingua Network service
EN ع
Research

Research Papers

Key research and technical papers from the major AI labs, summarized with their most important points.

All sources Anthropic OpenAI Meta AI Google DeepMind DeepSeek Kimi (Moonshot) Qwen (Alibaba) Hugging Face

Research Papers

Research paper
Hugging Face 7 Aug 2026

GST-Bench: New Benchmark Reveals VLMs' Global Spatial Awareness Gap

Hugging Face researchers introduce GST-Bench, a video-based benchmark for global spatial awareness, showing that the best zero-sho...

13
Research paper
Hugging Face 7 Aug 2026

Hugging Face Researchers Shrink MEG Speech Decoder 20x While Boosting Interpretability

A new study from Hugging Face presents a compact MEG-to-speech retrieval model that is 20 times smaller than prior systems yet ach...

14
Research paper
Hugging Face 7 Aug 2026

OSReward: Standardizing Evaluation of VLM Judges for Computer-Use Agents

A new benchmark, OSReward, systematically evaluates VLM judges on computer-use agent trajectories, revealing a systematic leniency...

14
Research paper
Hugging Face 7 Aug 2026

Hugging Face Study: On-Policy Delta Distillation Boosts Multilingual Math Reasoning

A new paper from Hugging Face explores On-Policy Delta Distillation (OPD²) for math reasoning in English, Korean, and Japanese, sh...

17
Research paper
Hugging Face 7 Aug 2026

EnvACE: Training LLM Agents by Rehearsing World Dynamics, Not Interacting with Environments

Hugging Face researchers propose EnvACE, a reinforcement learning method that lets LLM agents rehearse environment responses inter...

18
Research paper
Hugging Face 7 Aug 2026

ChronoVision: New Framework Boosts Temporal Reasoning in Multimodal AI

Hugging Face researchers introduce ChronoVision, a multimodal framework that improves temporal reasoning by reconstructing latent...

16
Research paper
Hugging Face 7 Aug 2026

UniME-R1: Learning from Failures with Retrieval-Centric Chain-of-Thought for Multimodal Retrieval

Hugging Face researchers introduce UniME-R1, an embedder-adviser framework that generates Retrieval-Centric Chain-of-Thought (RC-C...

13
Research paper
Hugging Face 7 Aug 2026

WorldClaw: Agentic Framework Generates Large-Scale 3D Worlds from Text

Hugging Face researchers introduce WorldClaw, an agentic coarse-to-fine framework that generates large-scale, editable 3D open wor...

11
Research paper
Hugging Face 7 Aug 2026

AgentOPSD: Recursive Self-Distillation Boosts Agentic RL Credit Assignment

Hugging Face researchers introduce AgentOPSD, a critic-free recursive method for turn-level credit assignment in agentic reinforce...

12
Research paper
Hugging Face 6 Aug 2026

WorldCycle: Self-Verifiable RL Cuts Drift in Video World Models by 44%

Hugging Face researchers introduce WorldCycle, a self-verifiable reinforcement learning framework that uses reversible action cycl...

14
Research paper
Hugging Face 6 Aug 2026

FocusMem: A New Latent Memory Framework for GUI Agents

Hugging Face researchers introduce FocusMem, a latent memory interface that separates content, readout, and trust to improve GUI a...

13
Research paper
Hugging Face 6 Aug 2026

Study: VLM Agents' Spatial Memory Goes Stale, Causing Safety Failures

A new empirical study reveals that memory-augmented VLM agents often fail to detect when their spatial memory is stale, leading to...

14
21 22 23 24 25