✨Stop the Flip-Flop: Context-Preserving Verification for Fast Revocable Diffusion Decoding

📝 Summary:
COVER stops flip-flop oscillations in parallel diffusion decoding with cache override verification. It performs leave-one-out verification and stable drafting in one pass, preserving context via KV cache override. This greatly reduces revisions for faster, quality-preserving decoding.

🔹 Publication Date: Published on Feb 5

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2602.06161
• PDF: https://arxiv.org/pdf/2602.06161

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

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

#DiffusionModels #GenerativeAI #DeepLearning #Decoding #ContextPreservation

✨LLMs Encode Their Failures: Predicting Success from Pre-Generation Activations

📝 Summary:
LLMs encode their likelihood of success in pre-generation activations. Probes can predict performance on math and coding tasks, outperforming surface features. This allows efficient inference routing across models, reducing costs by up to 70% while improving performance.

🔹 Publication Date: Published on Feb 10

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2602.09924
• PDF: https://arxiv.org/pdf/2602.09924
• Github: https://github.com/KabakaWilliam/llms_know_difficulty

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

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

#LLMs #AIResearch #MachineLearning #PerformancePrediction #CostEfficiency

✨Learning on the Manifold: Unlocking Standard Diffusion Transformers with Representation Encoders

📝 Summary:
Standard diffusion transformers fail on representation encoders due to geometric interference. Our RJF method uses Riemannian flow matching to guide generation along the manifold, enabling standard DiT architectures to converge effectively without width scaling.

🔹 Publication Date: Published on Feb 10

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2602.10099
• PDF: https://arxiv.org/pdf/2602.10099
• Github: https://github.com/amandpkr/RJF

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

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

#DiffusionModels #MachineLearning #GenerativeAI #ManifoldLearning #AIResearch

✨Learning to Continually Learn via Meta-learning Agentic Memory Designs

📝 Summary:
ALMA uses meta-learning to automatically discover adaptable memory designs for agentic systems, enabling continual learning without human engineering. Its learned designs outperform state-of-the-art human-crafted methods across diverse domains.

🔹 Publication Date: Published on Feb 8

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2602.07755
• PDF: https://arxiv.org/pdf/2602.07755
• Project Page: https://yimingxiong.me/alma
• Github: https://github.com/zksha/alma

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

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

#AI #DataScience #MachineLearning #HuggingFace #Research

✨TokenTrim: Inference-Time Token Pruning for Autoregressive Long Video Generation

📝 Summary:
Auto-regressive video generation suffers from temporal drift due to error accumulation in latent conditioning tokens, which is addressed by identifying and removing unstable tokens during inference to...

🔹 Publication Date: Published on Jan 30

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2602.00268
• PDF: https://arxiv.org/pdf/2602.00268
• Project Page: https://arielshaulov.github.io/TokenTrim/
• Github: https://github.com/arielshaulov/TokenTrim

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

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

#AI #DataScience #MachineLearning #HuggingFace #Research

✨C-ΔΘ: Circuit-Restricted Weight Arithmetic for Selective Refusal

📝 Summary:
Offline selective refusal in large language models is achieved through circuit-restricted weight updates that eliminate runtime intervention costs while maintaining performance. AI-generated summary M...

🔹 Publication Date: Published on Feb 4

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2602.04521
• PDF: https://arxiv.org/pdf/2602.04521

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

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

#AI #DataScience #MachineLearning #HuggingFace #Research

✨Bridging Academia and Industry: A Comprehensive Benchmark for Attributed Graph Clustering

📝 Summary:
PyAGC presents a production-ready benchmark and library for attributed graph clustering that addresses limitations of current research through scalable, memory-efficient implementations and comprehens...

🔹 Publication Date: Published on Feb 9

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2602.08519
• PDF: https://arxiv.org/pdf/2602.08519
• Project Page: https://pyagc.readthedocs.io
• Github: https://github.com/Cloudy1225/PyAGC

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

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

#AI #DataScience #MachineLearning #HuggingFace #Research

✨On the Optimal Reasoning Length for RL-Trained Language Models

📝 Summary:
Length control methods in reinforcement learning-trained language models affect reasoning performance and computational efficiency, with optimal output lengths balancing these factors. AI-generated su...

🔹 Publication Date: Published on Feb 10

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2602.09591
• PDF: https://arxiv.org/pdf/2602.09591

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

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

#AI #DataScience #MachineLearning #HuggingFace #Research

✨VISTA-Bench: Do Vision-Language Models Really Understand Visualized Text as Well as Pure Text?

📝 Summary:
VISTA-Bench evaluates vision-language models' ability to understand visualized text versus pure-text queries, revealing significant performance gaps and sensitivity to rendering variations. AI-generat...

🔹 Publication Date: Published on Feb 4

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2602.04802
• PDF: https://arxiv.org/pdf/2602.04802
• Github: https://github.com/QingAnLiu/VISTA-Bench

✨ Datasets citing this paper:
• https://huggingface.co/datasets/liuqa/VISTA-Bench

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

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

#AI #DataScience #MachineLearning #HuggingFace #Research

✨pySLAM: An Open-Source, Modular, and Extensible Framework for SLAM

📝 Summary:
pySLAM is an open-source framework supporting Visual SLAM with monocular, stereo, and RGB-D cameras, incorporating classical and modern features, loop closure methods, volumetric reconstruction, and d...

🔹 Publication Date: Published on Feb 17, 2025

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2502.11955
• PDF: https://arxiv.org/pdf/2502.11955
• Github: https://github.com/luigifreda/pyslam

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

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

#AI #DataScience #MachineLearning #HuggingFace #Research

✨Large-Scale Terminal Agentic Trajectory Generation from Dockerized Environments

📝 Summary:
A scalable pipeline called TerminalTraj addresses challenges in creating high-quality terminal trajectories for training agentic models by filtering repositories, generating Docker-aligned task instan...

🔹 Publication Date: Published on Feb 1

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2602.01244
• PDF: https://arxiv.org/pdf/2602.01244
• Github: https://github.com/multimodal-art-projection/TerminalTraj

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

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

#AI #DataScience #MachineLearning #HuggingFace #Research

✨LatentLens: Revealing Highly Interpretable Visual Tokens in LLMs

📝 Summary:
LatentLens enables interpretation of visual token representations in vision-language models by comparing them to contextualized textual representations, revealing that visual tokens are more interpret...

🔹 Publication Date: Published on Jan 31

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2602.00462
• PDF: https://arxiv.org/pdf/2602.00462
• Github: https://github.com/McGill-NLP/latentlens

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

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

#AI #DataScience #MachineLearning #HuggingFace #Research

✨Surprisal-Guided Selection: Compute-Optimal Test-Time Strategies for Execution-Grounded Code Generation

📝 Summary:
Test-time training fails in verification-grounded tasks due to over-sharpening, while surprisal-guided selection improves performance by favoring diverse, low-confidence samples. AI-generated summary ...

🔹 Publication Date: Published on Feb 7

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2602.07670
• PDF: https://arxiv.org/pdf/2602.07670
• Project Page: https://jbarnes850.github.io/2026/02/02/surprisal-guided-selection/
• Github: https://jbarnes850.github.io/2026/02/02/surprisal-guided-selection/

🔹 Models citing this paper:
• https://huggingface.co/Jarrodbarnes/KernelBench-RLVR-120b

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

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

#AI #DataScience #MachineLearning #HuggingFace #Research

✨Effective Reasoning Chains Reduce Intrinsic Dimensionality

📝 Summary:
Effective chain-of-thought reasoning strategies reduce intrinsic dimensionality, leading to better generalization by requiring fewer model parameters to achieve given accuracy thresholds. AI-generated...

🔹 Publication Date: Published on Feb 9

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2602.09276
• PDF: https://arxiv.org/pdf/2602.09276

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

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

#AI #DataScience #MachineLearning #HuggingFace #Research

✨ContextBench: A Benchmark for Context Retrieval in Coding Agents

📝 Summary:
ContextBench evaluates context retrieval in coding agents through detailed process analysis, revealing that advanced agent designs provide limited improvements in context usage while highlighting gaps...

🔹 Publication Date: Published on Feb 5

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2602.05892
• PDF: https://arxiv.org/pdf/2602.05892
• Project Page: https://contextbench.github.io/
• Github: https://github.com/EuniAI/ContextBench

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

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

#AI #DataScience #MachineLearning #HuggingFace #Research

✨Secure Code Generation via Online Reinforcement Learning with Vulnerability Reward Model

📝 Summary:
SecCoderX uses online reinforcement learning to align large language models for secure code generation while preserving functionality, addressing the functionality-security trade-off through vulnerabi...

🔹 Publication Date: Published on Feb 7

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2602.07422
• PDF: https://arxiv.org/pdf/2602.07422
• Github: https://github.com/AndrewWTY/SecCoderX

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

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

#AI #DataScience #MachineLearning #HuggingFace #Research

✨Contact-Anchored Policies: Contact Conditioning Creates Strong Robot Utility Models

📝 Summary:
Contact-Anchored Policies CAP replace language conditioning with physical contact points, using modular utility models for robust robot manipulation. CAP achieves superior zero-shot performance with minimal demonstration data, outperforming large VLAs by 56%.

🔹 Publication Date: Published on Feb 9

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2602.09017
• PDF: https://arxiv.org/pdf/2602.09017
• Project Page: https://cap-policy.github.io/
• Github: https://github.com/jeffacce/cap-policy

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

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

#AI #DataScience #MachineLearning #HuggingFace #Research

✨iGRPO: Self-Feedback-Driven LLM Reasoning

📝 Summary:
iGRPO enhances LLM mathematical reasoning using a two-stage, self-feedback process. It first drafts solutions, selects the best, and then refines based on that best draft. This iterative approach significantly improves performance and achieves state-of-the-art results on math benchmarks.

🔹 Publication Date: Published on Feb 9

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2602.09000
• PDF: https://arxiv.org/pdf/2602.09000

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

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

#AI #DataScience #MachineLearning #HuggingFace #Research

✨CausalArmor: Efficient Indirect Prompt Injection Guardrails via Causal Attribution

📝 Summary:
CausalArmor is a selective defense against Indirect Prompt Injection in AI agents. It uses causal ablation to detect when untrusted content dominates an agents privileged actions, triggering targeted sanitization only then. This improves security utility and latency over always-on defenses.

🔹 Publication Date: Published on Feb 8

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2602.07918
• PDF: https://arxiv.org/pdf/2602.07918

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

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

#AI #DataScience #MachineLearning #HuggingFace #Research

✨AgentSys: Secure and Dynamic LLM Agents Through Explicit Hierarchical Memory Management

📝 Summary:
AgentSys defends against indirect prompt injection in LLM agents through hierarchical memory isolation and controlled data flow, significantly reducing attack success rates while maintaining performan...

🔹 Publication Date: Published on Feb 7

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2602.07398
• PDF: https://arxiv.org/pdf/2602.07398
• Github: https://github.com/ruoyaow/agentsys-memory

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

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

#AI #DataScience #MachineLearning #HuggingFace #Research

✨Locas: Your Models are Principled Initializers of Locally-Supported Parametric Memories

📝 Summary:
Locas, a locally-supported parametric memory mechanism, enables flexible integration with transformer models for continual learning while minimizing catastrophic forgetting through principled initiali...

🔹 Publication Date: Published on Feb 4

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2602.05085
• PDF: https://arxiv.org/pdf/2602.05085

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

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

#AI #DataScience #MachineLearning #HuggingFace #Research