🔹 Title: BrowseComp-Plus: A More Fair and Transparent Evaluation Benchmark of Deep-Research Agent

🔹 Publication Date: Published on Aug 8

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2508.06600
• PDF: https://arxiv.org/pdf/2508.06600
• Project Page: https://texttron.github.io/BrowseComp-Plus/
• Github: https://github.com/texttron/BrowseComp-Plus

🔹 Datasets citing this paper:
• https://huggingface.co/datasets/Tevatron/browsecomp-plus-corpus
• https://huggingface.co/datasets/Tevatron/browsecomp-plus

🔹 Spaces citing this paper:
• https://huggingface.co/spaces/Tevatron/BrowseComp-Plus
==================================

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

🔹 Title: OmniEAR: Benchmarking Agent Reasoning in Embodied Tasks

🔹 Publication Date: Published on Aug 7

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2508.05614
• PDF: https://arxiv.org/pdf/2508.05614
• Project Page: https://zju-real.github.io/OmniEmbodied/
• Github: https://zju-real.github.io/OmniEmbodied/

🔹 Datasets citing this paper:
• https://huggingface.co/datasets/wangzx1210/OmniEAR

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

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

🔹 Title: MolmoAct: Action Reasoning Models that can Reason in Space

🔹 Publication Date: Published on Aug 11

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2508.07917
• PDF: https://arxiv.org/pdf/2508.07917

🔹 Datasets citing this paper:
No datasets found

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

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

🔹 Title: Temporal Self-Rewarding Language Models: Decoupling Chosen-Rejected via Past-Future

🔹 Publication Date: Published on Aug 8

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2508.06026
• PDF: https://arxiv.org/pdf/2508.06026

🔹 Datasets citing this paper:
No datasets found

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

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

🔹 Title: Reinforcement Learning in Vision: A Survey

🔹 Publication Date: Published on Aug 11

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2508.08189
• PDF: https://arxiv.org/pdf/2508.08189
• Github: https://github.com/weijiawu/Awesome-Visual-Reinforcement-Learning

🔹 Datasets citing this paper:
No datasets found

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

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

🔹 Title: SONAR-LLM: Autoregressive Transformer that Thinks in Sentence Embeddings and Speaks in Tokens

🔹 Publication Date: Published on Aug 7

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2508.05305
• PDF: https://arxiv.org/pdf/2508.05305
• Github: https://github.com/FusionBrainLab/SONAR-LLM/tree/main

🔹 Datasets citing this paper:
No datasets found

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

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

🔹 Title: Part I: Tricks or Traps? A Deep Dive into RL for LLM Reasoning

🔹 Publication Date: Published on Aug 11

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2508.08221
• PDF: https://arxiv.org/pdf/2508.08221

🔹 Datasets citing this paper:
No datasets found

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

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

🔹 Title: Follow-Your-Shape: Shape-Aware Image Editing via Trajectory-Guided Region Control

🔹 Publication Date: Published on Aug 11

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2508.08134
• PDF: https://arxiv.org/pdf/2508.08134
• Github: https://github.com/mayuelala/FollowYourShape

🔹 Datasets citing this paper:
No datasets found

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

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

🔹 Title: Less Is More: Training-Free Sparse Attention with Global Locality for Efficient Reasoning

🔹 Publication Date: Published on Aug 9

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2508.07101
• PDF: https://arxiv.org/pdf/2508.07101
• Github: https://github.com/DerrickYLJ/LessIsMore

🔹 Datasets citing this paper:
No datasets found

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

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

🔹 Title: VisR-Bench: An Empirical Study on Visual Retrieval-Augmented Generation for Multilingual Long Document Understanding

🔹 Publication Date: Published on Aug 10

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2508.07493
• PDF: https://arxiv.org/pdf/2508.07493
• Github: https://github.com/puar-playground/VisR-Bench

🔹 Datasets citing this paper:
• https://huggingface.co/datasets/puar-playground/VisR-Bench

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

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

🔹 Title: GLiClass: Generalist Lightweight Model for Sequence Classification Tasks

🔹 Publication Date: Published on Aug 11

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2508.07662
• PDF: https://arxiv.org/pdf/2508.07662

🔹 Datasets citing this paper:
No datasets found

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

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

🔹 Title: Deep Ignorance: Filtering Pretraining Data Builds Tamper-Resistant Safeguards into Open-Weight LLMs

🔹 Publication Date: Published on Aug 8

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2508.06601
• PDF: https://arxiv.org/pdf/2508.06601

🔹 Datasets citing this paper:
• https://huggingface.co/datasets/EleutherAI/deep-ignorance-pretraining-mix
• https://huggingface.co/datasets/EleutherAI/deep-ignorance-annealing-mix

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

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

🔹 Title: Bifrost-1: Bridging Multimodal LLMs and Diffusion Models with Patch-level CLIP Latents

🔹 Publication Date: Published on Aug 8

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2508.05954
• PDF: https://arxiv.org/pdf/2508.05954
• Project Page: https://bifrost-1.github.io/
• Github: https://github.com/HL-hanlin/Bifrost-1

🔹 Datasets citing this paper:
No datasets found

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

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

🔹 Title: When Good Sounds Go Adversarial: Jailbreaking Audio-Language Models with Benign Inputs

🔹 Publication Date: Published on Aug 5

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2508.03365
• PDF: https://arxiv.org/pdf/2508.03365

🔹 Datasets citing this paper:
No datasets found

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

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

🔹 Title: Speech-to-LaTeX: New Models and Datasets for Converting Spoken Equations and Sentences

🔹 Publication Date: Published on Aug 5

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2508.03542
• PDF: https://arxiv.org/pdf/2508.03542

🔹 Datasets citing this paper:
No datasets found

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

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

🔹 Title: Grove MoE: Towards Efficient and Superior MoE LLMs with Adjugate Experts

🔹 Publication Date: Published on Aug 11

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2508.07785
• PDF: https://arxiv.org/pdf/2508.07785
• Github: https://github.com/inclusionAI/GroveMoE/

🔹 Datasets citing this paper:
No datasets found

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

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

🔹 Title: A Comprehensive Survey of Self-Evolving AI Agents: A New Paradigm Bridging Foundation Models and Lifelong Agentic Systems

🔹 Publication Date: Published on Aug 10

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2508.07407
• PDF: https://arxiv.org/pdf/2508.07407

🔹 Datasets citing this paper:
No datasets found

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

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

🔹 Title: MoBE: Mixture-of-Basis-Experts for Compressing MoE-based LLMs

🔹 Publication Date: Published on Aug 7

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2508.05257
• PDF: https://arxiv.org/pdf/2508.05257

🔹 Datasets citing this paper:
No datasets found

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

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

🔹 Title: Goedel-Prover-V2: Scaling Formal Theorem Proving with Scaffolded Data Synthesis and Self-Correction

🔹 Publication Date: Published on Aug 5

🔹 Abstract: Goedel-Prover-V2, a series of open-source language models, achieves state-of-the-art performance in automated theorem proving through scaffolded data synthesis, verifier-guided self-correction, and model averaging. AI-generated summary We introduce Goedel-Prover-V2, a series of open-source language models that set a new state-of-the-art in automated theorem proving . Built on the standard expert iteration and reinforcement learning pipeline, our approach incorporates three key innovations: (1) Scaffolded data synthesis : We generate synthetic tasks of increasing difficulty to train the model to master increasingly complex theorems; (2) Verifier-guided self-correction : We enable the model to iteratively revise its proofs by leveraging feedback from the Lean compiler; (3) Model averaging : We merge model checkpoints to mitigate the decrease in model output diversity in later stages of training. Our small model, Goedel-Prover-V2-8B, reaches 84.6% pass@32 on MiniF2F and outperforms DeepSeek-Prover-V2-671B under the same metric, despite being 80X smaller. Our flagship model, Goedel-Prover-V2-32B, achieves 88.1% on MiniF2F at pass@32 in standard mode and 90.4% in self-correction mode, outperforming prior SOTA by a large margin. Additionally, our flagship model solves 86 problems on PutnamBench at pass@184 , securing the first place among open-source models on the leaderboard, surpassing DeepSeek-Prover-V2-671B's record of solving 47 problems by pass@1024 with a significantly smaller model size and compute budget. At the time of its release (July-August 2025), Goedel-Prover-V2 achieves the strongest overall performance among all open-source theorem provers. It also ranks among the top-performing models--including closed-source systems with publicly reported performance--under a constrained test-time compute budget. Our models, code, and data are released at https://github.com/Goedel-LM/Goedel-Prover-V2.

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2508.03613

• PDF: https://arxiv.org/pdf/2508.03613

🔹 Datasets citing this paper:
No datasets found

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

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

🔹 Title: Compressing Chain-of-Thought in LLMs via Step Entropy

🔹 Publication Date: Published on Aug 5

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2508.03346
• PDF: https://arxiv.org/pdf/2508.03346

🔹 Datasets citing this paper:
No datasets found

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

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