🔹 Title: Large Language Models Do NOT Really Know What They Don't Know

🔹 Publication Date: Published on Oct 10

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2510.09033
• PDF: https://arxiv.org/pdf/2510.09033

🔹 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: BitNet Distillation

🔹 Publication Date: Published on Oct 15

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2510.13998
• PDF: https://arxiv.org/pdf/2510.13998
• Github: https://github.com/microsoft/BitNet

🔹 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: TokDrift: When LLM Speaks in Subwords but Code Speaks in Grammar

🔹 Publication Date: Published on Oct 16

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2510.14972
• PDF: https://arxiv.org/pdf/2510.14972
• Github: https://github.com/uw-swag/tokdrift

🔹 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: Fantastic (small) Retrievers and How to Train Them: mxbai-edge-colbert-v0 Tech Report

🔹 Publication Date: Published on Oct 16

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2510.14880
• PDF: https://arxiv.org/pdf/2510.14880

🔹 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: RefusalBench: Generative Evaluation of Selective Refusal in Grounded Language Models

🔹 Publication Date: Published on Oct 12

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2510.10390
• PDF: https://arxiv.org/pdf/2510.10390

🔹 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: Expertise need not monopolize: Action-Specialized Mixture of Experts for Vision-Language-Action Learning

🔹 Publication Date: Published on Oct 16

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2510.14300
• PDF: https://arxiv.org/pdf/2510.14300

🔹 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: LiteStage: Latency-aware Layer Skipping for Multi-stage Reasoning

🔹 Publication Date: Published on Oct 16

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2510.14211
• PDF: https://arxiv.org/pdf/2510.14211

🔹 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: Ponimator: Unfolding Interactive Pose for Versatile Human-human Interaction Animation

🔹 Publication Date: Published on Oct 16

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2510.14976
• PDF: https://arxiv.org/pdf/2510.14976
• Project Page: https://stevenlsw.github.io/ponimator/
• Github: https://github.com/stevenlsw/ponimator

🔹 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: COIG-Writer: A High-Quality Dataset for Chinese Creative Writing with Thought Processes

🔹 Publication Date: Published on Oct 16

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2510.14763
• PDF: https://arxiv.org/pdf/2510.14763
• Project Page: https://huggingface.co/m-a-p
• Github: https://COIG-Writer.github.io/

🔹 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: Beyond Correctness: Evaluating Subjective Writing Preferences Across Cultures

🔹 Publication Date: Published on Oct 16

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2510.14616
• PDF: https://arxiv.org/pdf/2510.14616
• Github: https://WritingPreferenceBench.github.io/

🔹 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 German Commons - 154 Billion Tokens of Openly Licensed Text for German Language Models

🔹 Publication Date: Published on Oct 15

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2510.13996
• PDF: https://arxiv.org/pdf/2510.13996

🔹 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: RealDPO: Real or Not Real, that is the Preference

🔹 Publication Date: Published on Oct 16

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2510.14955
• PDF: https://arxiv.org/pdf/2510.14955
• Project Page: https://vchitect.github.io/RealDPO-Project/
• Github: https://github.com/Vchitect/RealDPO

🔹 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: DialectGen: Benchmarking and Improving Dialect Robustness in Multimodal Generation

🔹 Publication Date: Published on Oct 16

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2510.14949
• PDF: https://arxiv.org/pdf/2510.14949
• Project Page: https://dialectgen.github.io/
• Github: https://github.com/DialectGen/DialectGen

🔹 Datasets citing this paper:
• https://huggingface.co/datasets/uclanlp/DialectGen

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

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

🔹 Title: VLA^2: Empowering Vision-Language-Action Models with an Agentic Framework for Unseen Concept Manipulation

🔹 Publication Date: Published on Oct 16

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2510.14902
• PDF: https://arxiv.org/pdf/2510.14902
• Project Page: https://vla-2.github.io

🔹 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: SCas4D: Structural Cascaded Optimization for Boosting Persistent 4D Novel View Synthesis

🔹 Publication Date: Published on Oct 8

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2510.06694
• PDF: https://arxiv.org/pdf/2510.06694

🔹 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: ImagerySearch: Adaptive Test-Time Search for Video Generation Beyond Semantic Dependency Constraints

🔹 Publication Date: Published on Oct 16

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2510.14847
• PDF: https://arxiv.org/pdf/2510.14847
• Github: https://github.com/AMAP-ML/ImagerySearch

🔹 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: RAGCap-Bench: Benchmarking Capabilities of LLMs in Agentic Retrieval Augmented Generation Systems

🔹 Publication Date: Published on Oct 15

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2510.13910
• PDF: https://arxiv.org/pdf/2510.13910

🔹 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: When Models Lie, We Learn: Multilingual Span-Level Hallucination Detection with PsiloQA

🔹 Publication Date: Published on Oct 6

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2510.04849
• PDF: https://arxiv.org/pdf/2510.04849

🔹 Datasets citing this paper:
• https://huggingface.co/datasets/s-nlp/PsiloQA

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

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

🔹 Title: On Pretraining for Project-Level Code Completion

🔹 Publication Date: Published on Oct 15

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2510.13697
• PDF: https://arxiv.org/pdf/2510.13697
• Project Page: https://huggingface.co/collections/JetBrains-Research/repository-level-pre-trained-opencoder-68e938c003be1cfba9c3595e

🔹 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

🤖🧠 NVIDIA, MIT, HKU and Tsinghua University Introduce QeRL: A Powerful Quantum Leap in Reinforcement Learning for LLMs

🗓️ 17 Oct 2025
📚 AI News & Trends

The rise of large language models (LLMs) has redefined artificial intelligence powering everything from conversational AI to autonomous reasoning systems. However, training these models especially through reinforcement learning (RL) is computationally expensive requiring massive GPU resources and long training cycles. To address this, a team of researchers from NVIDIA, Massachusetts Institute of Technology (MIT), The ...

#QuantumLearning #ReinforcementLearning #LLMs #NVIDIA #MIT #TsinghuaUniversity

🔹 Title: Efficient Parallel Samplers for Recurrent-Depth Models and Their Connection to Diffusion Language Models

🔹 Publication Date: Published on Oct 16

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2510.14961
• PDF: https://arxiv.org/pdf/2510.14961
• Github: https://github.com/seal-rg/recurrent-pretraining

🔹 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