ML Research Hub – Telegram
ML Research Hub
32.8K subscribers
4.4K photos
272 videos
23 files
4.76K links
Advancing research in Machine Learning – practical insights, tools, and techniques for researchers.

Admin: @HusseinSheikho || @Hussein_Sheikho
Download Telegram
🔹 Title: VOGUE: Guiding Exploration with Visual Uncertainty Improves Multimodal Reasoning

🔹 Publication Date: Published on Oct 1

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2510.01444
• PDF: https://arxiv.org/pdf/2510.01444

🔹 Datasets citing this paper:
No datasets found

🔹 Spaces citing this paper:
No spaces found
==================================

For more data science resources:
https://news.1rj.ru/str/DataScienceT
🔹 Title: RLP: Reinforcement as a Pretraining Objective

🔹 Publication Date: Published on Sep 26

🔹 Paper Links:
• arXiv Page: https://arxiv.org/pdf/2510.01265
• PDF: https://arxiv.org/pdf/2510.01265
• Project Page: https://t.co/6PNJYfiAoJ
• Github: https://github.com/NVlabs/RLP

🔹 Datasets citing this paper:
No datasets found

🔹 Spaces citing this paper:
No spaces found
==================================

For more data science resources:
https://news.1rj.ru/str/DataScienceT
🔹 Title: The Unreasonable Effectiveness of Scaling Agents for Computer Use

🔹 Publication Date: Published on Oct 2

🔹 Paper Links:
• arXiv Page: http://arxiv.org/abs/2510.02250
• PDF: https://arxiv.org/pdf/2510.02250
• Project Page: https://www.simular.ai/articles/agent-s3
• Github: http://github.com/simular-ai/Agent-S

🔹 Datasets citing this paper:
No datasets found

🔹 Spaces citing this paper:
No spaces found
==================================

For more data science resources:
https://news.1rj.ru/str/DataScienceT
1
🔹 Title: F2LLM Technical Report: Matching SOTA Embedding Performance with 6 Million Open-Source Data

🔹 Publication Date: Published on Oct 2

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2510.02294
• PDF: https://arxiv.org/pdf/2510.02294
• Github: https://github.com/codefuse-ai/CodeFuse-Embeddings/tree/main/F2LLM

🔹 Datasets citing this paper:
https://huggingface.co/datasets/codefuse-ai/F2LLM

🔹 Spaces citing this paper:
No spaces found
==================================

For more data science resources:
https://news.1rj.ru/str/DataScienceT
🔹 Title: StockBench: Can LLM Agents Trade Stocks Profitably In Real-world Markets?

🔹 Publication Date: Published on Oct 2

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2510.02209
• PDF: https://arxiv.org/pdf/2510.02209

🔹 Datasets citing this paper:
No datasets found

🔹 Spaces citing this paper:
No spaces found
==================================

For more data science resources:
https://news.1rj.ru/str/DataScienceT
🔹 Title: A Rigorous Benchmark with Multidimensional Evaluation for Deep Research Agents: From Answers to Reports

🔹 Publication Date: Published on Oct 2

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2510.02190
• PDF: https://arxiv.org/pdf/2510.02190

🔹 Datasets citing this paper:
No datasets found

🔹 Spaces citing this paper:
No spaces found
==================================

For more data science resources:
https://news.1rj.ru/str/DataScienceT
🔹 Title: Group-Relative REINFORCE Is Secretly an Off-Policy Algorithm: Demystifying Some Myths About GRPO and Its Friends

🔹 Publication Date: Published on Sep 29

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2509.24203
• PDF: https://arxiv.org/pdf/2509.24203
• Github: https://github.com/modelscope/Trinity-RFT/tree/main/examples/rec_gsm8k

🔹 Datasets citing this paper:
No datasets found

🔹 Spaces citing this paper:
No spaces found
==================================

For more data science resources:
https://news.1rj.ru/str/DataScienceT
🔹 Title: SKYLENAGE Technical Report: Mathematical Reasoning and Contest-Innovation Benchmarks for Multi-Level Math Evaluation

🔹 Publication Date: Published on Sep 24

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2510.01241
• PDF: https://arxiv.org/pdf/2510.01241

🔹 Datasets citing this paper:
No datasets found

🔹 Spaces citing this paper:
No spaces found
==================================

For more data science resources:
https://news.1rj.ru/str/DataScienceT
🔹 Title: Parallel Scaling Law: Unveiling Reasoning Generalization through A Cross-Linguistic Perspective

🔹 Publication Date: Published on Oct 2

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2510.02272
• PDF: https://arxiv.org/pdf/2510.02272

🔹 Datasets citing this paper:
No datasets found

🔹 Spaces citing this paper:
No spaces found
==================================

For more data science resources:
https://news.1rj.ru/str/DataScienceT
🔹 Title: Just Do It!? Computer-Use Agents Exhibit Blind Goal-Directedness

🔹 Publication Date: Published on Oct 2

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2510.01670
• PDF: https://arxiv.org/pdf/2510.01670

🔹 Datasets citing this paper:
No datasets found

🔹 Spaces citing this paper:
No spaces found
==================================

For more data science resources:
https://news.1rj.ru/str/DataScienceT
🔹 Title: Automated Structured Radiology Report Generation with Rich Clinical Context

🔹 Publication Date: Published on Oct 1

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2510.00428
• PDF: https://arxiv.org/pdf/2510.00428
• Github: https://github.com/vuno/contextualized-srrg

🔹 Datasets citing this paper:
No datasets found

🔹 Spaces citing this paper:
No spaces found
==================================

For more data science resources:
https://news.1rj.ru/str/DataScienceT
🔹 Title: Go with Your Gut: Scaling Confidence for Autoregressive Image Generation

🔹 Publication Date: Published on Sep 30

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2509.26376
• PDF: https://arxiv.org/pdf/2509.26376
• Github: https://github.com/EnVision-Research/ScalingAR

🔹 Datasets citing this paper:
No datasets found

🔹 Spaces citing this paper:
No spaces found
==================================

For more data science resources:
https://news.1rj.ru/str/DataScienceT
🔹 Title: FrameThinker: Learning to Think with Long Videos via Multi-Turn Frame Spotlighting

🔹 Publication Date: Published on Sep 29

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2509.24304
• PDF: https://arxiv.org/pdf/2509.24304
• Project Page: https://github.com/lcqysl/FrameThinker-RL
• Github: https://github.com/lcqysl/FrameThinker-RL

🔹 Datasets citing this paper:
No datasets found

🔹 Spaces citing this paper:
No spaces found
==================================

For more data science resources:
https://news.1rj.ru/str/DataScienceT
🔹 Title: VLA-R1: Enhancing Reasoning in Vision-Language-Action Models

🔹 Publication Date: Published on Oct 2

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2510.01623
• PDF: https://arxiv.org/pdf/2510.01623
• Project Page: https://gigaai-research.github.io/VLA-R1
• Github: https://github.com/GigaAI-research/VLA-R1

🔹 Datasets citing this paper:
No datasets found

🔹 Spaces citing this paper:
No spaces found
==================================

For more data science resources:
https://news.1rj.ru/str/DataScienceT
🔹 Title: RewardMap: Tackling Sparse Rewards in Fine-grained Visual Reasoning via Multi-Stage Reinforcement Learning

🔹 Publication Date: Published on Oct 2

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2510.02240
• PDF: https://arxiv.org/pdf/2510.02240
• Project Page: https://fscdc.github.io/RewardMap
• Github: https://fscdc.github.io/RewardMap

🔹 Datasets citing this paper:
No datasets found

🔹 Spaces citing this paper:
No spaces found
==================================

For more data science resources:
https://news.1rj.ru/str/DataScienceT
🔹 Title: Optimal Control Meets Flow Matching: A Principled Route to Multi-Subject Fidelity

🔹 Publication Date: Published on Oct 2

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2510.02315
• PDF: https://arxiv.org/pdf/2510.02315
• Github: https://github.com/ericbill21/FOCUS/

🔹 Datasets citing this paper:
No datasets found

🔹 Spaces citing this paper:
No spaces found
==================================

For more data science resources:
https://news.1rj.ru/str/DataScienceT
🔹 Title: ExGRPO: Learning to Reason from Experience

🔹 Publication Date: Published on Oct 2

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2510.02245
• PDF: https://arxiv.org/pdf/2510.02245

🔹 Datasets citing this paper:
No datasets found

🔹 Spaces citing this paper:
No spaces found
==================================

For more data science resources:
https://news.1rj.ru/str/DataScienceT
🔹 Title: TimeSeriesScientist: A General-Purpose AI Agent for Time Series Analysis

🔹 Publication Date: Published on Oct 2

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2510.01538
• PDF: https://arxiv.org/pdf/2510.01538

🔹 Datasets citing this paper:
No datasets found

🔹 Spaces citing this paper:
No spaces found
==================================

For more data science resources:
https://news.1rj.ru/str/DataScienceT
🔹 Title: Spectral Scaling Laws in Language Models: How Effectively Do Feed-Forward Networks Use Their Latent Space?

🔹 Publication Date: Published on Oct 1

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2510.00537
• PDF: https://arxiv.org/pdf/2510.00537

🔹 Datasets citing this paper:
No datasets found

🔹 Spaces citing this paper:
No spaces found
==================================

For more data science resources:
https://news.1rj.ru/str/DataScienceT
🔹 Title: Agentic Jigsaw Interaction Learning for Enhancing Visual Perception and Reasoning in Vision-Language Models

🔹 Publication Date: Published on Oct 1

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2510.01304
• PDF: https://arxiv.org/pdf/2510.01304
• Project Page: https://yuzeng0-0.github.io/AGILE/
• Github: https://github.com/yuzeng0-0/AGILE

🔹 Datasets citing this paper:
No datasets found

🔹 Spaces citing this paper:
No spaces found
==================================

For more data science resources:
https://news.1rj.ru/str/DataScienceT
🔹 Title: Learning to Reason for Hallucination Span Detection

🔹 Publication Date: Published on Oct 2

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2510.02173
• PDF: https://arxiv.org/pdf/2510.02173

🔹 Datasets citing this paper:
No datasets found

🔹 Spaces citing this paper:
No spaces found
==================================

For more data science resources:
https://news.1rj.ru/str/DataScienceT