🔹 Title: RL-PLUS: Countering Capability Boundary Collapse of LLMs in Reinforcement Learning with Hybrid-policy Optimization

🔹 Publication Date: Published on Jul 31

🔹 Abstract: RL-PLUS, a hybrid-policy optimization approach, enhances LLM reasoning capabilities by integrating Multiple Importance Sampling and Exploration-Based Advantage Function, outperforming RLVR on various benchmarks and resolving capability boundary collapse. AI-generated summary Reinforcement Learning with Verifiable Reward (RLVR) has significantly advanced the complex reasoning abilities of Large Language Models (LLMs) . However, it struggles to break through the inherent capability boundaries of the base LLM, due to its essentially on-policy strategy coupled with LLM's immense action space and sparse reward. Critically, RLVR can lead to the capability boundary collapse , narrowing the LLM's problem-solving scope. To address this problem, we propose RL-PLUS, a novel hybrid-policy optimization approach for LLMs that synergizes internal exploitation with external data to achieve stronger reasoning capabilities and surpass the boundaries of base models. RL-PLUS integrates two core components, i.e., Multiple Importance Sampling to address distributional mismatch from external data, and Exploration-Based Advantage Function to guide the model towards high-value, unexplored reasoning paths. We provide both theoretical analysis and extensive experiments to demonstrate the superiority and generalizability of our approach. Compared with existing RLVR methods, RL-PLUS achieves 1) state-of-the-art performance on six math reasoning benchmarks ; 2) superior performance on six out-of-distribution reasoning tasks ; 3) consistent and significant gains across diverse model families, with average relative improvements up to 69.2\%. Moreover, the analysis of Pass@k curves indicates that RL-PLUS effectively resolves the capability boundary collapse problem.

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2508.00222

• PDF: https://arxiv.org/pdf/2508.00222

🔹 Datasets citing this paper:
No datasets found

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

For more data science resources:
✓ https://t.me/DataScienceT

🔹 Title: TempFlow-GRPO: When Timing Matters for GRPO in Flow Models

🔹 Publication Date: Published on Aug 6

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2508.04324
• PDF: https://arxiv.org/pdf/2508.04324
• Project Page: https://tempflowgrpo.github.io/
• Github: https://github.com/Shredded-Pork/TempFlow-GRPO

🔹 Datasets citing this paper:
No datasets found

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

For more data science resources:
✓ https://t.me/DataScienceT

🔹 Title: Advances in Speech Separation: Techniques, Challenges, and Future Trends

🔹 Publication Date: Published on Aug 14

🔹 Paper Links:
• arXiv Page: https://arxiv.org/pdf/2508.10830
• PDF: https://arxiv.org/pdf/2508.10830
• Project Page: https://cslikai.cn/Speech-Separation-Paper-Tutorial
• Github: https://github.com/JusperLee/Speech-Separation-Paper-Tutorial

🔹 Datasets citing this paper:
No datasets found

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

For more data science resources:
✓ https://t.me/DataScienceT

🔹 Title: Copyright Protection for Large Language Models: A Survey of Methods, Challenges, and Trends

🔹 Publication Date: Published on Aug 15

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2508.11548
• PDF: https://arxiv.org/pdf/2508.11548
• Github: https://xuzhenhua55.github.io/awesome-llm-copyright-protection/index.html

🔹 Datasets citing this paper:
No datasets found

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

For more data science resources:
✓ https://t.me/DataScienceT

🔹 Title: CorrSteer: Steering Improves Task Performance and Safety in LLMs through Correlation-based Sparse Autoencoder Feature Selection

🔹 Publication Date: Published on Aug 18

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2508.12535
• PDF: https://arxiv.org/pdf/2508.12535

🔹 Datasets citing this paper:
No datasets found

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

For more data science resources:
✓ https://t.me/DataScienceT

🔹 Title: Radiance Fields in XR: A Survey on How Radiance Fields are Envisioned and Addressed for XR Research

🔹 Publication Date: Published on Aug 6

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2508.04326
• PDF: https://arxiv.org/pdf/2508.04326
• Project Page: https://mediated-reality.github.io/rf4xr/papers/li_tvcg25/
• Github: https://github.com/mediated-reality/awesome-rf4xr

🔹 Datasets citing this paper:
No datasets found

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

For more data science resources:
✓ https://t.me/DataScienceT

🔹 Title: ZARA: Zero-shot Motion Time-Series Analysis via Knowledge and Retrieval Driven LLM Agents

🔹 Publication Date: Published on Aug 6

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2508.04038
• PDF: https://arxiv.org/pdf/2508.04038
• Github: https://github.com/zechenli03/ZARA

🔹 Datasets citing this paper:
No datasets found

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

For more data science resources:
✓ https://t.me/DataScienceT

🔹 Title: Evaluating Podcast Recommendations with Profile-Aware LLM-as-a-Judge

🔹 Publication Date: Published on Aug 12

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2508.08777
• PDF: https://arxiv.org/pdf/2508.08777

🔹 Datasets citing this paper:
No datasets found

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

For more data science resources:
✓ https://t.me/DataScienceT

🔹 Title: MedSAMix: A Training-Free Model Merging Approach for Medical Image Segmentation

🔹 Publication Date: Published on Aug 14

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2508.11032
• PDF: https://arxiv.org/pdf/2508.11032
• Github: https://github.com/podismine/MedSAMix

🔹 Datasets citing this paper:
No datasets found

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

For more data science resources:
✓ https://t.me/DataScienceT

🔹 Title: Semantic IDs for Joint Generative Search and Recommendation

🔹 Publication Date: Published on Aug 14

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2508.10478
• PDF: https://arxiv.org/pdf/2508.10478

🔹 Datasets citing this paper:
No datasets found

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

For more data science resources:
✓ https://t.me/DataScienceT

🔹 Title: Describe What You See with Multimodal Large Language Models to Enhance Video Recommendations

🔹 Publication Date: Published on Aug 13

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2508.09789
• PDF: https://arxiv.org/pdf/2508.09789

🔹 Datasets citing this paper:
• https://huggingface.co/datasets/marcodena/video-recs-describe-what-you-see

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

For more data science resources:
✓ https://t.me/DataScienceT

🔹 Title: Embodied-R1: Reinforced Embodied Reasoning for General Robotic Manipulation

🔹 Publication Date: Published on Aug 19

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2508.13998
• PDF: https://arxiv.org/pdf/2508.13998

🔹 Datasets citing this paper:
No datasets found

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

For more data science resources:
✓ https://t.me/DataScienceT

🔹 Title: Beyond Human Judgment: A Bayesian Evaluation of LLMs' Moral Values Understanding

🔹 Publication Date: Published on Aug 19

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2508.13804
• PDF: https://arxiv.org/pdf/2508.13804
• Project Page: https://maciejskorski.github.io/moral-foundations-llm-eval
• Github: https://github.com/maciejskorski/moral-foundations-llm-eval

🔹 Datasets citing this paper:
No datasets found

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

For more data science resources:
✓ https://t.me/DataScienceT

🔹 Title: Mind the Generation Process: Fine-Grained Confidence Estimation During LLM Generation

🔹 Publication Date: Published on Aug 16

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2508.12040
• PDF: https://arxiv.org/pdf/2508.12040

🔹 Datasets citing this paper:
No datasets found

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

For more data science resources:
✓ https://t.me/DataScienceT

🔹 Title: A Stitch in Time Saves Nine: Proactive Self-Refinement for Language Models

🔹 Publication Date: Published on Aug 18

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2508.12903
• PDF: https://arxiv.org/pdf/2508.12903

🔹 Datasets citing this paper:
No datasets found

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

For more data science resources:
✓ https://t.me/DataScienceT

🔹 Title: MM-BrowseComp: A Comprehensive Benchmark for Multimodal Browsing Agents

🔹 Publication Date: Published on Aug 14

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2508.13186
• PDF: https://arxiv.org/pdf/2508.13186
• Github: https://github.com/MMBrowseComp/MM-BrowseComp

🔹 Datasets citing this paper:
No datasets found

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

For more data science resources:
✓ https://t.me/DataScienceT

🔹 Title: CAMAR: Continuous Actions Multi-Agent Routing

🔹 Publication Date: Published on Aug 18

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2508.12845
• PDF: https://arxiv.org/pdf/2508.12845

🔹 Datasets citing this paper:
No datasets found

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

For more data science resources:
✓ https://t.me/DataScienceT

🔹 Title: Atom-Searcher: Enhancing Agentic Deep Research via Fine-Grained Atomic Thought Reward

🔹 Publication Date: Published on Aug 18

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2508.12800
• PDF: https://arxiv.org/pdf/2508.12800

🔹 Datasets citing this paper:
No datasets found

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

For more data science resources:
✓ https://t.me/DataScienceT

🔥 $10.000 WITH LISA!

Lisa earned $200,000 in a month, and now it’s YOUR TURN!

She’s made trading SO SIMPLE that anyone can do it.

❗️Just copy her signals every day
❗️Follow her trades step by step
❗️Earn $1,000+ in your first week – GUARANTEED!

🚨 BONUS: Lisa is giving away $10,000 to her subscribers!

Don’t miss this once-in-a-lifetime opportunity. Free access for the first 500 people only!

👉 CLICK HERE TO JOIN NOW 👈

🔹 Title: Llama-3.1-FoundationAI-SecurityLLM-8B-Instruct Technical Report

🔹 Publication Date: Published on Aug 1

🔹 Abstract: Foundation-Sec-8B-Instruct is a cybersecurity-focused LLM designed for chat-style interactions and instruction-following, outperforming other models in cybersecurity tasks while matching their instruction-following capabilities. AI-generated summary Large language models ( LLMs ) have shown remarkable success across many domains, yet their integration into cybersecurity applications remains limited due to a lack of general-purpose cybersecurity data, representational complexity, and safety and regulatory concerns. To address this gap, we previously introduced Foundation-Sec-8B , a cybersecurity -focused LLM suitable for fine-tuning on downstream tasks. That model, however, was not designed for chat-style interactions or instruction-following . In this report, we release Foundation-Sec-8B -Instruct: a model specifically trained for general-purpose cybersecurity dialogue . Built on Foundation-Sec-8B , it combines domain-specific knowledge with instruction-following , conversational capabilities , and alignment with human preferences to produce high-quality, relevant responses. Comprehensive evaluations show that Foundation-Sec-8B -Instruct outperforms Llama 3.1-8B-Instruct on a range of cybersecurity tasks while matching its instruction-following performance. It is also competitive with GPT-4o-mini on cyber threat intelligence and instruction-following tasks. We envision Foundation-Sec-8B -Instruct becoming an indispensable assistant in the daily workflows of cybersecurity professionals. We release the model publicly at https://huggingface.co/fdtn-ai/ Foundation-Sec-8B -Instruct.

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2508.01059

• PDF: https://arxiv.org/pdf/2508.01059

🔹 Datasets citing this paper:
No datasets found

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

For more data science resources:
✓ https://t.me/DataScienceT

🔹 Title: Rapidly Adapting to New Voice Spoofing: Few-Shot Detection of Synthesized Speech Under Distribution Shifts

🔹 Publication Date: Published on Aug 18

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2508.13320
• PDF: https://arxiv.org/pdf/2508.13320

🔹 Datasets citing this paper:
No datasets found

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

For more data science resources:
✓ https://t.me/DataScienceT