✨Complementary Reinforcement Learning

📝 Summary:
Complementary RL enables efficient agent learning by synchronizing experience extraction with policy optimization through dual objectives that evolve together during training. AI-generated summary Rei...

🔹 Publication Date: Published on Mar 18

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2603.17621
• PDF: https://arxiv.org/pdf/2603.17621
• Github: https://github.com/pUmpKin-Co/ComplementaryRL

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

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

#AI #DataScience #MachineLearning #HuggingFace #Research

✨MetaClaw: Just Talk -- An Agent That Meta-Learns and Evolves in the Wild

📝 Summary:
MetaClaw is a continual meta-learning framework for LLM agents that evolves policies and reusable skills. It enables zero-downtime skill adaptation and opportunistic policy optimization during inactive periods. This boosts agent accuracy and robustness, scaling to production LLMs.

🔹 Publication Date: Published on Mar 17

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2603.17187
• PDF: https://arxiv.org/pdf/2603.17187
• Github: https://github.com/aiming-lab/MetaClaw

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

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

#AI #DataScience #MachineLearning #HuggingFace #Research

✨Efficient Exploration at Scale

📝 Summary:
An online learning algorithm for reinforcement learning from human feedback that achieves significant data efficiency improvements through incremental model updates, reward uncertainty modeling, and i...

🔹 Publication Date: Published on Mar 18

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2603.17378
• PDF: https://arxiv.org/pdf/2603.17378

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

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

#AI #DataScience #MachineLearning #HuggingFace #Research

✨LoST: Level of Semantics Tokenization for 3D Shapes

📝 Summary:
Level-of-Semantics Tokenization (LoST) improves 3D shape generation by ordering tokens based on semantic salience and using a novel relational alignment loss for better reconstruction and efficiency. ...

🔹 Publication Date: Published on Mar 18

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2603.17995
• PDF: https://arxiv.org/pdf/2603.17995
• Project Page: https://lost3d.github.io/

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

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

#AI #DataScience #MachineLearning #HuggingFace #Research

✨RAMP: Reinforcement Adaptive Mixed Precision Quantization for Efficient On Device LLM Inference

📝 Summary:
Reinforcement learning-based mixed precision quantization method achieves superior compression efficiency and model performance for large language models through adaptive bit width assignment and nove...

🔹 Publication Date: Published on Mar 18

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2603.17891
• PDF: https://arxiv.org/pdf/2603.17891

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

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

#AI #DataScience #MachineLearning #HuggingFace #Research

✨PRISM: Demystifying Retention and Interaction in Mid-Training

📝 Summary:
Mid-training design choices significantly improve reasoning performance in large language models, with optimal results achieved when reinforcement learning is applied to models that have been pre-trai...

🔹 Publication Date: Published on Mar 17

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2603.17074
• PDF: https://arxiv.org/pdf/2603.17074

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

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

#AI #DataScience #MachineLearning #HuggingFace #Research

✨ESPIRE: A Diagnostic Benchmark for Embodied Spatial Reasoning of Vision-Language Models

📝 Summary:
ESPIRE is a diagnostic benchmark for embodied spatial reasoning that evaluates vision-language models on robotic tasks through a decomposed localization and execution framework, enabling fine-grained ...

🔹 Publication Date: Published on Mar 13

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2603.13033
• PDF: https://arxiv.org/pdf/2603.13033
• Project Page: https://spatigen.github.io/espire.io/
• Github: https://github.com/spatigen/espire

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

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

#AI #DataScience #MachineLearning #HuggingFace #Research

✨Temporal Gains, Spatial Costs: Revisiting Video Fine-Tuning in Multimodal Large Language Models

📝 Summary:
Video-supervised fine-tuning in multimodal large language models consistently enhances video performance while often degrading static image benchmarks, with frame sampling frequency determining the ex...

🔹 Publication Date: Published on Mar 18

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2603.17541
• PDF: https://arxiv.org/pdf/2603.17541

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

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

#AI #DataScience #MachineLearning #HuggingFace #Research

✨Conservative Offline Robot Policy Learning via Posterior-Transition Reweighting

📝 Summary:
Posterior-Transition Reweighting (PTR) improves offline robot policy adaptation by dynamically weighting training samples based on the attribution of their post-action consequences, enabling more cons...

🔹 Publication Date: Published on Mar 17

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2603.16542
• PDF: https://arxiv.org/pdf/2603.16542
• Project Page: https://research.beingbeyond.com/ptr
• Github: https://github.com/BeingBeyond/PTR

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

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

#AI #DataScience #MachineLearning #HuggingFace #Research

✨VideoAtlas: Navigating Long-Form Video in Logarithmic Compute

📝 Summary:
VideoAtlas enables lossless video representation through hierarchical grids, enabling efficient long-context processing via recursive language models with adaptive compute allocation. AI-generated sum...

🔹 Publication Date: Published on Mar 18

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2603.17948
• PDF: https://arxiv.org/pdf/2603.17948
• Project Page: https://mohammad2012191.github.io/VideoAtlas/
• Github: https://github.com/mohammad2012191/VideoAtlas

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

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

#AI #DataScience #MachineLearning #HuggingFace #Research

✨Efficient Training-Free Multi-Token Prediction via Embedding-Space Probing

📝 Summary:
A training-free method for multi-token prediction in large language models using mask tokens from the embedding space enables parallel token generation with improved throughput and accuracy. AI-genera...

🔹 Publication Date: Published on Mar 18

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2603.17942
• PDF: https://arxiv.org/pdf/2603.17942

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

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

#AI #DataScience #MachineLearning #HuggingFace #Research

✨BenchPreS: A Benchmark for Context-Aware Personalized Preference Selectivity of Persistent-Memory LLMs

📝 Summary:
LLMs struggle to apply user preferences context-sensitively, treating them as universal rules. BenchPreS evaluates this, showing even frontier LLMs over-apply preferences in third-party settings. This problem persists despite reasoning or prompt defenses.

🔹 Publication Date: Published on Mar 17

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2603.16557
• PDF: https://arxiv.org/pdf/2603.16557

✨ Datasets citing this paper:
• https://huggingface.co/datasets/sangyon/BenchPreS

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

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

#LLMs #Personalization #ContextAwareAI #AIResearch #Benchmarking

✨AI Scientist via Synthetic Task Scaling

📝 Summary:
This paper introduces a novel synthetic environment pipeline that generates machine learning challenges for training AI agents. Student models trained with these synthetic tasks, using teacher trajectories, achieve significantly improved performance on MLGym benchmarks.

🔹 Publication Date: Published on Mar 17

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2603.17216
• PDF: https://arxiv.org/pdf/2603.17216

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

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

#AIResearch #MachineLearning #SyntheticTasks #AIAgents #DeepLearning

✨Look Before Acting: Enhancing Vision Foundation Representations for Vision-Language-Action Models

📝 Summary:
VLA models struggle to integrate visual detail for action generation. DeepVision-VLA enhances visual representations via multi-level feature injection and action-guided pruning. This significantly boosts performance on robotic tasks.

🔹 Publication Date: Published on Mar 16

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2603.15618
• PDF: https://arxiv.org/pdf/2603.15618
• Project Page: https://deepvision-vla.github.io/

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

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

#VLAModels #ComputerVision #Robotics #DeepLearning #FoundationModels

✨GigaWorld-Policy: An Efficient Action-Centered World--Action Model

📝 Summary:
GigaWorld-Policy is an action-centered World-Action Model that significantly improves robotic policy learning. It decouples visual and motion representations, using dual supervision from action prediction and video generation. This allows for 9x faster inference and 7% higher task success rates c...

🔹 Publication Date: Published on Mar 18

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2603.17240
• PDF: https://arxiv.org/pdf/2603.17240
• Project Page: https://gigaai-research.github.io/GigaWorld-Policy/
• Github: https://github.com/open-gigaai/giga-world-policy

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

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

#Robotics #MachineLearning #WorldModels #DeepLearning #PolicyLearning

✨Video-CoE: Reinforcing Video Event Prediction via Chain of Events

📝 Summary:
Video-CoE introduces a Chain of Events CoE paradigm to improve video event prediction. It addresses MLLM limitations in logical reasoning and visual utilization by constructing temporal event chains and using enhanced training. CoE achieves state-of-the-art performance on VEP benchmarks.

🔹 Publication Date: Published on Mar 16

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2603.14935
• PDF: https://arxiv.org/pdf/2603.14935

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

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

#VideoEventPrediction #ChainOfEvents #MLLM #ComputerVision #AI

✨Alignment Makes Language Models Normative, Not Descriptive

📝 Summary:
Aligned language models excel at normative, rule-based behavior prediction but struggle with complex descriptive human strategic interactions. Base models predict real human choices in these games better. This reveals a trade-off in model optimization.

🔹 Publication Date: Published on Mar 17

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2603.17218
• PDF: https://arxiv.org/pdf/2603.17218

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

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

#LLM #AIAlignment #NormativeAI #GameTheory #AIBehavior

✨ACE-LoRA: Graph-Attentive Context Enhancement for Parameter-Efficient Adaptation of Medical Vision-Language Models

📝 Summary:
ACE-LoRA parameter-efficiently adapts medical VLMs, enhancing zero-shot generalization. It integrates LoRA and attention-based context enhancement to capture fine-grained diagnostic cues. This outperforms state-of-the-art models across diverse medical tasks.

🔹 Publication Date: Published on Mar 17

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2603.17079
• PDF: https://arxiv.org/pdf/2603.17079
• Github: https://github.com/icon-lab/ACE-LoRA

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

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

#MedicalAI #VisionLanguageModels #LoRA #DeepLearning #EfficientAI

✨FINER: MLLMs Hallucinate under Fine-grained Negative Queries

📝 Summary:
Multimodal language models hallucinate under fine-grained negative queries, a gap in existing benchmarks. This paper introduces FINER benchmarks and FINER-Tuning, a DPO method, to address this. It significantly reduces hallucinations and boosts general MLLM capabilities.

🔹 Publication Date: Published on Mar 18

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2603.17662
• PDF: https://arxiv.org/pdf/2603.17662
• Project Page: https://explainableml.github.io/finer-project/
• Github: https://github.com/ExplainableML/finer

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

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

#MLLMs #AIHallucinations #Benchmarking #DeepLearning #AIResearch

✨HeBA: Heterogeneous Bottleneck Adapters for Robust Vision-Language Models

📝 Summary:
HeBA introduces a heterogeneous bottleneck adapter framework for Vision-Language Models. It uses modality-specific processing like convolutions for images and linear projections for text, combined with a compression bottleneck and active gradient initialization. This design improves few-shot lear...

🔹 Publication Date: Published on Mar 17

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2603.16653
• PDF: https://arxiv.org/pdf/2603.16653
• Project Page: https://huggingface.co/papers?q=dense%20linear%20projections
• Github: https://github.com/Jahid12012021/VLM-HeBA

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

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

#VisionLanguageModels #DeepLearning #AIResearch #ModelAdapters #FewShotLearning

✨Coherent Human-Scene Reconstruction from Multi-Person Multi-View Video in a Single Pass

📝 Summary:
CHROMM is a unified framework that jointly reconstructs cameras, scene point clouds, and human meshes from multi-person multi-view videos. It integrates strong priors, handles scale discrepancies, and uses multi-view fusion for faster, more robust human-scene reconstruction.

🔹 Publication Date: Published on Mar 13

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2603.12789
• PDF: https://arxiv.org/pdf/2603.12789
• Project Page: https://nstar1125.github.io/chromm
• Github: https://nstar1125.github.io/chromm/

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

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

#3DReconstruction #ComputerVision #HumanSceneReconstruction #MultiViewVideo #AIResearch