✨Agent S2: A Compositional Generalist-Specialist Framework for Computer Use Agents

📝 Summary:
Agent S2 is a new compositional framework for computer use agents. It uses Mixture-of-Grounding and Proactive Hierarchical Planning to achieve state-of-the-art performance across various benchmarks and operating systems, significantly improving automation.

🔹 Publication Date: Published on Apr 1, 2025

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2504.00906
• PDF: https://arxiv.org/pdf/2504.00906
• Project Page: https://www.simular.ai/articles/agent-s2-technical-review
• Github: https://github.com/simular-ai/Agent-S

==================================

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

#AI #AIagents #Automation #MachineLearning #ComputerScience

✨MAEB: Massive Audio Embedding Benchmark

📝 Summary:
MAEB is a large-scale audio benchmark evaluating 50+ models across 30 diverse tasks. No single model dominates; strengths vary significantly between speech and environmental sound tasks. Performance on MAEB highly correlates with audio large language model performance.

🔹 Publication Date: Published on Feb 17

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2602.16008
• PDF: https://arxiv.org/pdf/2602.16008
• Project Page: https://mteb-leaderboard.hf.space/?benchmark_name=MAEB%28beta%29

==================================

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

#AudioAI #Benchmarking #AudioEmbeddings #SpeechProcessing #AudioLLMs

✨CADEvolve: Creating Realistic CAD via Program Evolution

📝 Summary:
CADEvolve presents an evolution-based pipeline using VLM-guided edits to generate complex CAD programs from simple primitives. It creates a large dataset of 1.3 million scripts, enabling fine-tuned VLMs to achieve state-of-the-art Image2CAD performance.

🔹 Publication Date: Published on Feb 18

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2602.16317
• PDF: https://arxiv.org/pdf/2602.16317
• Github: https://github.com/zhemdi/CADEvolve

🔹 Models citing this paper:
• https://huggingface.co/kulibinai/cadevolve-rl1

✨ Datasets citing this paper:
• https://huggingface.co/datasets/kulibinai/cadevolve

==================================

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

#CAD #ProgramEvolution #VLMs #Image2CAD #AI

✨Reinforced Fast Weights with Next-Sequence Prediction

📝 Summary:
REFINE is an RL framework that improves fast weight models for long-context tasks. It uses next-sequence prediction NSP instead of next-token prediction, enhancing long-range dependency capture. Experiments show it consistently outperforms supervised fine-tuning.

🔹 Publication Date: Published on Feb 18

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2602.16704
• PDF: https://arxiv.org/pdf/2602.16704
• Github: https://github.com/princetonvisualai/ReFINE/

==================================

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

#ReinforcementLearning #MachineLearning #DeepLearning #NLP #AI

✨Learning Personalized Agents from Human Feedback

📝 Summary:
PAHF enables AI agents to continually personalize through explicit user memory and dual feedback. It rapidly adapts to changing user preferences by integrating pre-action clarification and post-action updates, significantly reducing personalization error and improving learning speed.

🔹 Publication Date: Published on Feb 18

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2602.16173
• PDF: https://arxiv.org/pdf/2602.16173
• Project Page: https://personalized-ai.github.io/
• Github: https://github.com/facebookresearch/PAHF

==================================

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

#AI #Personalization #HumanAIInteraction #MachineLearning #AIAgents

✨OPBench: A Graph Benchmark to Combat the Opioid Crisis

📝 Summary:
OPBench is the first comprehensive graph benchmark to systematically evaluate graph learning methods for the opioid crisis. It includes five diverse datasets across three domains and a unified framework, providing insights for future research.

🔹 Publication Date: Published on Feb 16

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2602.14602
• PDF: https://arxiv.org/pdf/2602.14602
• Github: https://github.com/Tianyi-Billy-Ma/OPBench

==================================

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

#OpioidCrisis #GraphLearning #DataScience #MachineLearning #PublicHealth

✨Unified Latents (UL): How to train your latents

📝 Summary:
Unified Latents UL learns joint latent representations using diffusion prior regularization and decoding. It achieves competitive FID of 1.4 on ImageNet-512 with fewer training FLOPs and sets a new state of the art FVD of 1.3 on Kinetics-600.

🔹 Publication Date: Published on Feb 19

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2602.17270
• PDF: https://arxiv.org/pdf/2602.17270

==================================

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

#GenerativeAI #DiffusionModels #LatentSpace #ImageGeneration #VideoGeneration

✨Arcee Trinity Large Technical Report

📝 Summary:
Arcee Trinity introduces sparse Mixture-of-Experts models Nano, Mini, Large with up to 400B total parameters. They feature advanced attention, novel normalization, and sigmoid MoE routing, trained on massive token datasets.

🔹 Publication Date: Published on Feb 19

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2602.17004
• PDF: https://arxiv.org/pdf/2602.17004

==================================

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

#MixtureOfExperts #LargeLanguageModels #SparseModels #DeepLearning #AI

✨TactAlign: Human-to-Robot Policy Transfer via Tactile Alignment

📝 Summary:
TactAlign transfers human tactile demonstrations to robots with different embodiments. It aligns human and robot tactile signals into a shared latent space without paired data, improving policy transfer for contact-rich tasks and enabling zero-shot transfer.

🔹 Publication Date: Published on Feb 14

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2602.13579
• PDF: https://arxiv.org/pdf/2602.13579
• Project Page: https://yswi.github.io/tactalign/

==================================

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

#Robotics #TactileRobotics #PolicyTransfer #HRI #AI

✨Frontier AI Risk Management Framework in Practice: A Risk Analysis Technical Report v1.5

📝 Summary:
This report assesses frontier AI risks, updating granular scenarios for cyber offense, manipulation, deception, uncontrolled AI R&D, and self-replication. It also proposes robust mitigation strategies for secure deployment of advanced AI systems.

🔹 Publication Date: Published on Feb 16

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2602.14457
• PDF: https://arxiv.org/pdf/2602.14457
• Project Page: https://ai45lab.github.io/safeworkf1-page/

==================================

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

#AI #DataScience #MachineLearning #HuggingFace #Research

✨SpargeAttention2: Trainable Sparse Attention via Hybrid Top-k+Top-p Masking and Distillation Fine-Tuning

📝 Summary:
A trainable sparse attention method called SpargeAttention2 is proposed that achieves high sparsity in diffusion models while maintaining generation quality through hybrid masking rules and distillati...

🔹 Publication Date: Published on Feb 13

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2602.13515
• PDF: https://arxiv.org/pdf/2602.13515

==================================

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

#AI #DataScience #MachineLearning #HuggingFace #Research

✨Mobile-Agent-v3.5: Multi-platform Fundamental GUI Agents

📝 Summary:
GUI-Owl-1.5 is a multi-platform GUI agent model with varying sizes that achieves superior performance across GUI automation, grounding, tool-calling, and memory tasks through innovations in data pipel...

🔹 Publication Date: Published on Feb 15

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2602.16855
• PDF: https://arxiv.org/pdf/2602.16855
• Project Page: https://github.com/X-PLUG/MobileAgent/tree/main/Mobile-Agent-v3.5
• Github: https://github.com/X-PLUG/MobileAgent

==================================

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

#AI #DataScience #MachineLearning #HuggingFace #Research

✨DDiT: Dynamic Patch Scheduling for Efficient Diffusion Transformers

📝 Summary:
Dynamic tokenization improves diffusion transformer efficiency by adjusting patch sizes based on content complexity and denoising timestep, achieving significant speedup without quality loss. AI-gener...

🔹 Publication Date: Published on Feb 19

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2602.16968
• PDF: https://arxiv.org/pdf/2602.16968

==================================

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

#AI #DataScience #MachineLearning #HuggingFace #Research

✨Computer-Using World Model

📝 Summary:
A world model for desktop software that predicts UI state changes through textual description followed by visual synthesis, improving decision quality and execution robustness in computer-using tasks....

🔹 Publication Date: Published on Feb 19

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2602.17365
• PDF: https://arxiv.org/pdf/2602.17365

==================================

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

#AI #DataScience #MachineLearning #HuggingFace #Research

✨2Mamba2Furious: Linear in Complexity, Competitive in Accuracy

📝 Summary:
Researchers enhance linear attention by simplifying Mamba-2 and improving its architectural components to achieve near-softmax accuracy while maintaining memory efficiency for long sequences. AI-gener...

🔹 Publication Date: Published on Feb 19

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2602.17363
• PDF: https://arxiv.org/pdf/2602.17363
• Github: https://github.com/gmongaras/2Mamba2Furious

==================================

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

#AI #DataScience #MachineLearning #HuggingFace #Research

✨Discovering Multiagent Learning Algorithms with Large Language Models

📝 Summary:
AlphaEvolve, an evolutionary coding agent using large language models, automatically discovers new multiagent learning algorithms for imperfect-information games by evolving regret minimization and po...

🔹 Publication Date: Published on Feb 18

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2602.16928
• PDF: https://arxiv.org/pdf/2602.16928

==================================

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

#AI #DataScience #MachineLearning #HuggingFace #Research

✨PaperBench: Evaluating AI's Ability to Replicate AI Research

📝 Summary:
PaperBench evaluates AI agents' ability to replicate state-of-the-art AI research by decomposing replication tasks into graded sub-tasks, using both LLM-based and human judges to assess performance. A...

🔹 Publication Date: Published on Apr 2, 2025

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2504.01848
• PDF: https://arxiv.org/pdf/2504.01848
• Github: https://github.com/openai/preparedness

✨ Datasets citing this paper:
• https://huggingface.co/datasets/josancamon/paperbench
• https://huggingface.co/datasets/ai-coscientist/researcher-ablation-bench

==================================

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

#AI #DataScience #MachineLearning #HuggingFace #Research

✨References Improve LLM Alignment in Non-Verifiable Domains

📝 Summary:
References improve LLM alignment in non-verifiable domains. Reference-guided LLM-evaluators act as soft verifiers, boosting judge accuracy and enabling self-improvement for post-training. This method outperforms SFT and reference-free techniques, achieving strong results.

🔹 Publication Date: Published on Feb 18

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2602.16802
• PDF: https://arxiv.org/pdf/2602.16802
• Github: https://github.com/yale-nlp/RLRR

==================================

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

#AI #DataScience #MachineLearning #HuggingFace #Research

✨FRAPPE: Infusing World Modeling into Generalist Policies via Multiple Future Representation Alignment

📝 Summary:
FRAPPE addresses limitations in world modeling for robotics by using parallel progressive expansion to improve representation alignment and reduce error accumulation in predictive models. AI-generated...

🔹 Publication Date: Published on Feb 19

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2602.17259
• PDF: https://arxiv.org/pdf/2602.17259
• Project Page: https://h-zhao1997.github.io/frappe/
• Github: https://github.com/OpenHelix-Team/frappe

==================================

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

#AI #DataScience #MachineLearning #HuggingFace #Research

✨"What Are You Doing?": Effects of Intermediate Feedback from Agentic LLM In-Car Assistants During Multi-Step Processing

📝 Summary:
Intermediate feedback from in-car AI assistants improves user experience, trust, and perceived speed, reducing task load. Users prefer adaptive feedback, starting transparently and becoming less verbose as reliability increases.

🔹 Publication Date: Published on Feb 17

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2602.15569
• PDF: https://arxiv.org/pdf/2602.15569
• Github: https://github.com/johanneskirmayr/agentic_llm_feedback

==================================

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

#LLM #AI #HCI #AutomotiveAI #UserExperience

✨World Models for Policy Refinement in StarCraft II

📝 Summary:
StarWM is the first world model for StarCraft II predicting future observations under partial observability using a structured textual representation. It achieves significant offline prediction accuracy and, integrated into a decision system, yields substantial win-rate improvements against SC2s ...

🔹 Publication Date: Published on Feb 16

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2602.14857
• PDF: https://arxiv.org/pdf/2602.14857
• Github: https://github.com/yxzzhang/StarWM

🔹 Models citing this paper:
• https://huggingface.co/yxzhang2024/StarWM

✨ Datasets citing this paper:
• https://huggingface.co/datasets/yxzhang2024/SC2-Dynamics-50K

==================================

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

#WorldModels #StarCraftII #AI #ReinforcementLearning #DeepLearning