ML Research Hub – Telegram
ML Research Hub
32.8K subscribers
4.38K photos
270 videos
23 files
4.74K links
Advancing research in Machine Learning – practical insights, tools, and techniques for researchers.

Admin: @HusseinSheikho || @Hussein_Sheikho
Download Telegram
🔹 Title: An Empirical Study of Testing Practices in Open Source AI Agent Frameworks and Agentic Applications

🔹 Publication Date: Published on Sep 23

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2509.19185
• PDF: https://arxiv.org/pdf/2509.19185

🔹 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: DeepSearch: Overcome the Bottleneck of Reinforcement Learning with Verifiable Rewards via Monte Carlo Tree Search

🔹 Publication Date: Published on Sep 29

🔹 Paper Links:
• arXiv Page: https://arxiv.org/pdf/2509.25454
• PDF: https://arxiv.org/pdf/2509.25454

🔹 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: Infusing Theory of Mind into Socially Intelligent LLM Agents

🔹 Publication Date: Published on Sep 26

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2509.22887
• PDF: https://arxiv.org/pdf/2509.22887

🔹 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: Knapsack RL: Unlocking Exploration of LLMs via Optimizing Budget Allocation

🔹 Publication Date: Published on Sep 30

🔹 Paper Links:
• arXiv Page: https://www.arxiv.org/abs/2509.25849
• PDF: https://arxiv.org/pdf/2509.25849

🔹 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: Making, not Taking, the Best of N

🔹 Publication Date: Published on Oct 1

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2510.00931
• PDF: https://arxiv.org/pdf/2510.00931

🔹 Datasets citing this paper:
https://huggingface.co/datasets/CohereLabs/fusion-synth-data-geofactx
https://huggingface.co/datasets/CohereLabs/fusion-pairwise-evals-test-time-scaling
https://huggingface.co/datasets/CohereLabs/fusion-pairwise-evals-finetuned
https://huggingface.co/datasets/CohereLabs/fusion-synth-data-ufb

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

For more data science resources:
https://news.1rj.ru/str/DataScienceT
1
🔹 Title: BroRL: Scaling Reinforcement Learning via Broadened Exploration

🔹 Publication Date: Published on Oct 1

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2510.01180
• PDF: https://arxiv.org/pdf/2510.01180

🔹 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
2
🔹 Title: ACON: Optimizing Context Compression for Long-horizon LLM Agents

🔹 Publication Date: Published on Oct 1

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2510.00615
• PDF: https://arxiv.org/pdf/2510.00615

🔹 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: Eliciting Secret Knowledge from Language Models

🔹 Publication Date: Published on Oct 1

🔹 Paper Links:
• arXiv Page: https://huggingface.co/collections/bcywinski/eliciting-secret-knowledge-from-language-models-68de1a49ae6fa034e5c105ff
• PDF: https://arxiv.org/pdf/2510.01070
• Github: https://github.com/cywinski/eliciting-secret-knowledge

🔹 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: ReSWD: ReSTIR'd, not shaken. Combining Reservoir Sampling and Sliced Wasserstein Distance for Variance Reduction

🔹 Publication Date: Published on Oct 1

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2510.01061
• PDF: https://arxiv.org/pdf/2510.01061
• Project Page: https://reservoirswd.github.io/
• Github: https://reservoirswd.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: CurES: From Gradient Analysis to Efficient Curriculum Learning for Reasoning LLMs

🔹 Publication Date: Published on Oct 1

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2510.01037
• PDF: https://arxiv.org/pdf/2510.01037
• Github: https://github.com/ZexuSun/CurES

🔹 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: VLM-FO1: Bridging the Gap Between High-Level Reasoning and Fine-Grained Perception in VLMs

🔹 Publication Date: Published on Sep 30

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2509.25916
• PDF: https://arxiv.org/pdf/2509.25916

🔹 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: Hyperdimensional Probe: Decoding LLM Representations via Vector Symbolic Architectures

🔹 Publication Date: Published on Sep 29

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2509.25045
• PDF: https://arxiv.org/pdf/2509.25045
• Github: https://github.com/Ipazia-AI/hyperprobe

🔹 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: PIPer: On-Device Environment Setup via Online Reinforcement Learning

🔹 Publication Date: Published on Sep 29

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2509.25455
• PDF: https://arxiv.org/pdf/2509.25455
• Github: https://github.com/JetBrains-Research/PIPer

🔹 Datasets citing this paper:
https://huggingface.co/datasets/JetBrains-Research/PIPer-envbench-zeroshot-rl
https://huggingface.co/datasets/JetBrains-Research/PIPer-SFT-2500-sharegpt

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

For more data science resources:
https://news.1rj.ru/str/DataScienceT
🔹 Title: BindWeave: Subject-Consistent Video Generation via Cross-Modal Integration

🔹 Publication Date: Published on Oct 1

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2510.00438
• PDF: https://arxiv.org/pdf/2510.00438
• Project Page: https://lzy-dot.github.io/BindWeave/
• Github: https://lzy-dot.github.io/BindWeave

🔹 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
2
🔹 Title: It Takes Two: Your GRPO Is Secretly DPO

🔹 Publication Date: Published on Oct 1

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2510.00977
• PDF: https://arxiv.org/pdf/2510.00977

🔹 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: SINQ: Sinkhorn-Normalized Quantization for Calibration-Free Low-Precision LLM Weights

🔹 Publication Date: Published on Sep 26

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2509.22944
• PDF: https://arxiv.org/pdf/2509.22944
• Project Page: https://github.com/huawei-csl/SINQ
• Github: https://github.com/huawei-csl/SINQ

🔹 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: QUASAR: Quantum Assembly Code Generation Using Tool-Augmented LLMs via Agentic RL

🔹 Publication Date: Published on Oct 1

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2510.00967
• PDF: https://arxiv.org/pdf/2510.00967
• Github: https://github.com/benyucong/QUASAR

🔹 Datasets citing this paper:
https://huggingface.co/datasets/Benyucong/graph-data-quantum-rl

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

For more data science resources:
https://news.1rj.ru/str/DataScienceT
🔹 Title: EditReward: A Human-Aligned Reward Model for Instruction-Guided Image Editing

🔹 Publication Date: Published on Sep 30

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2509.26346
• PDF: https://arxiv.org/pdf/2509.26346

🔹 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: TGPO: Temporal Grounded Policy Optimization for Signal Temporal Logic Tasks

🔹 Publication Date: Published on Sep 30

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2510.00225
• PDF: https://arxiv.org/pdf/2510.00225

🔹 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: MixtureVitae: Open Web-Scale Pretraining Dataset With High Quality Instruction and Reasoning Data Built from Permissive-First Text Sources

🔹 Publication Date: Published on Sep 29

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2509.25531
• PDF: https://arxiv.org/pdf/2509.25531

🔹 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