ML Research Hub – Telegram
ML Research Hub
32.7K subscribers
4.08K photos
237 videos
23 files
4.4K links
Advancing research in Machine Learning – practical insights, tools, and techniques for researchers.

Admin: @HusseinSheikho || @Hussein_Sheikho
Download Telegram
🔹 Title: DivMerge: A divergence-based model merging method for multi-tasking

🔹 Publication Date: Published on Sep 2

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2509.02108
• PDF: https://arxiv.org/pdf/2509.02108

🔹 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: Mechanistic interpretability for steering vision-language-action models

🔹 Publication Date: Published on Aug 30

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2509.00328
• PDF: https://arxiv.org/pdf/2509.00328

🔹 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
4
🔹 Title: SFR-DeepResearch: Towards Effective Reinforcement Learning for Autonomously Reasoning Single Agents

🔹 Publication Date: Published on Sep 8

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2509.06283
• PDF: https://arxiv.org/pdf/2509.06283

🔹 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: Staying in the Sweet Spot: Responsive Reasoning Evolution via Capability-Adaptive Hint Scaffolding

🔹 Publication Date: Published on Sep 8

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2509.06923
• PDF: https://arxiv.org/pdf/2509.06923
• Github: https://github.com/ChillingDream/seele

🔹 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-R1: Towards Parallel Thinking via Reinforcement Learning

🔹 Publication Date: Published on Sep 9

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2509.07980
• PDF: https://arxiv.org/pdf/2509.07980
• Github: https://github.com/zhengkid/Parallel-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: Mini-o3: Scaling Up Reasoning Patterns and Interaction Turns for Visual Search

🔹 Publication Date: Published on Sep 9

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2509.07969
• PDF: https://arxiv.org/pdf/2509.07969
• Project Page: https://mini-o3.github.io/
• Github: https://github.com/Mini-o3/Mini-o3

🔹 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: Visual Representation Alignment for Multimodal Large Language Models

🔹 Publication Date: Published on Sep 9

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2509.07979
• PDF: https://arxiv.org/pdf/2509.07979
• Project Page: https://cvlab-kaist.github.io/VIRAL/

🔹 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: SimpleQA Verified: A Reliable Factuality Benchmark to Measure Parametric Knowledge

🔹 Publication Date: Published on Sep 9

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2509.07968
• PDF: https://arxiv.org/pdf/2509.07968
• Project Page: https://www.kaggle.com/benchmarks/deepmind/simpleqa-verified

🔹 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: F1: A Vision-Language-Action Model Bridging Understanding and Generation to Actions

🔹 Publication Date: Published on Sep 8

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2509.06951
• PDF: https://arxiv.org/pdf/2509.06951
• Github: https://github.com/InternRobotics/F1-VLA

🔹 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: Reconstruction Alignment Improves Unified Multimodal Models

🔹 Publication Date: Published on Sep 8

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2509.07295
• PDF: https://arxiv.org/pdf/2509.07295
• Project Page: https://reconstruction-alignment.github.io/
• Github: https://github.com/HorizonWind2004/reconstruction-alignment

🔹 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: Language Self-Play For Data-Free Training

🔹 Publication Date: Published on Sep 9

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2509.07414
• PDF: https://arxiv.org/pdf/2509.07414

🔹 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: UMO: Scaling Multi-Identity Consistency for Image Customization via Matching Reward

🔹 Publication Date: Published on Sep 8

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2509.06818
• PDF: https://arxiv.org/pdf/2509.06818
• Project Page: https://bytedance.github.io/UMO/
• Github: https://bytedance.github.io/UMO/

🔹 Datasets citing this paper:
No datasets found

🔹 Spaces citing this paper:
https://huggingface.co/spaces/bytedance-research/UMO_OmniGen2
https://huggingface.co/spaces/bytedance-research/UMO_UNO
==================================

For more data science resources:
https://news.1rj.ru/str/DataScienceT
🔹 Title: Causal Attention with Lookahead Keys

🔹 Publication Date: Published on Sep 9

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2509.07301
• PDF: https://arxiv.org/pdf/2509.07301

🔹 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: Q-Sched: Pushing the Boundaries of Few-Step Diffusion Models with Quantization-Aware Scheduling

🔹 Publication Date: Published on Sep 1

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2509.01624
• PDF: https://arxiv.org/pdf/2509.01624
• Github: https://github.com/enyac-group/q-sched

🔹 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: Curia: A Multi-Modal Foundation Model for Radiology

🔹 Publication Date: Published on Sep 8

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2509.06830
• PDF: https://arxiv.org/pdf/2509.06830
• Project Page: https://huggingface.co/raidium/curia

🔹 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: ΔL Normalization: Rethink Loss Aggregation in RLVR

🔹 Publication Date: Published on Sep 9

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2509.07558
• PDF: https://arxiv.org/pdf/2509.07558
• Github: https://github.com/zerolllin/Delta-L-Normalization

🔹 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: Directly Aligning the Full Diffusion Trajectory with Fine-Grained Human Preference

🔹 Publication Date: Published on Sep 8

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2509.06942
• PDF: https://arxiv.org/pdf/2509.06942

🔹 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: Emergent Hierarchical Reasoning in LLMs through Reinforcement Learning

🔹 Publication Date: Published on Sep 3

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2509.03646
• PDF: https://arxiv.org/pdf/2509.03646
• Project Page: https://tiger-ai-lab.github.io/Hierarchical-Reasoner/
• Github: https://github.com/TIGER-AI-Lab/Hierarchical-Reasoner

🔹 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: Benchmarking Information Retrieval Models on Complex Retrieval Tasks

🔹 Publication Date: Published on Sep 8

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2509.07253
• PDF: https://arxiv.org/pdf/2509.07253

🔹 Datasets citing this paper:
https://huggingface.co/datasets/jfkback/crumb

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

For more data science resources:
https://news.1rj.ru/str/DataScienceT
2
Forwarded from PyData Careers
حوّل فكرتك إلى مشروع ناجح مع ‎#سهول 🚀
منصّة عربية متكاملة لإدارة المشاريع تعتمد على الذكاء الاصطناعي، لتنظيم مهامك وفريقك بسهولة ومرونة.
جرّب النسخة التجريبية الآن مجاناً على: suhol.ai

للاستفسارات مجموعة سهول على تيليجرام: @suhol_ai

#إدارة_المشاريع#ذكاء_اصطناعي#kanban#مشاريع#تطوير
🔹 Title: From Noise to Narrative: Tracing the Origins of Hallucinations in Transformers

🔹 Publication Date: Published on Sep 8

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2509.06938
• PDF: https://arxiv.org/pdf/2509.06938

🔹 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