✨COMPASS: A Framework for Evaluating Organization-Specific Policy Alignment in LLMs

📝 Summary:
COMPASS evaluates large language models' compliance with organizational policies, revealing significant gaps in enforcing prohibitions despite strong performance on legitimate requests. AI-generated s...

🔹 Publication Date: Published on Jan 5

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2601.01836
• PDF: https://arxiv.org/pdf/2601.01836
• Github: https://github.com/AIM-Intelligence/COMPASS

🔹 Models citing this paper:
• https://huggingface.co/AIM-Intelligence/COMPASS_Qwen2.5-7B-Instruct_LoRA
• https://huggingface.co/AIM-Intelligence/COMPASS_gemma-3-4b-it_LoRA

✨ Datasets citing this paper:
• https://huggingface.co/datasets/AIM-Intelligence/COMPASS-Policy-Alignment-Testbed-Dataset
• https://huggingface.co/datasets/AIM-Intelligence/COMPASS-Policy-aware-SFT-Dataset

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

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

#AI #DataScience #MachineLearning #HuggingFace #Research

✨Project Ariadne: A Structural Causal Framework for Auditing Faithfulness in LLM Agents

📝 Summary:
Project Ariadne uses structural causal models and counterfactual logic to evaluate the causal integrity of LLM reasoning, revealing a faithfulness gap where reasoning traces are not reliable drivers o...

🔹 Publication Date: Published on Jan 5

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2601.02314
• PDF: https://arxiv.org/pdf/2601.02314
• Github: https://github.com/skhanzad/AridadneXAI

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

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

#AI #DataScience #MachineLearning #HuggingFace #Research

✨GARDO: Reinforcing Diffusion Models without Reward Hacking

📝 Summary:
Online reinforcement learning for diffusion model fine-tuning suffers from reward hacking due to proxy reward mismatches, which GARDO addresses through selective regularization, adaptive reference upd...

🔹 Publication Date: Published on Dec 30, 2025

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2512.24138
• PDF: https://arxiv.org/pdf/2512.24138
• Project Page: https://tinnerhrhe.github.io/gardo_project/
• Github: https://github.com/tinnerhrhe/gardo

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

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

#AI #DataScience #MachineLearning #HuggingFace #Research

✨IMA++: ISIC Archive Multi-Annotator Dermoscopic Skin Lesion Segmentation Dataset

📝 Summary:
A large-scale public multi-annotator skin lesion segmentation dataset is introduced with extensive metadata for annotator analysis and consensus modeling. AI-generated summary Multi-annotator medical ...

🔹 Publication Date: Published on Dec 25, 2025

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2512.21472
• PDF: https://arxiv.org/pdf/2512.21472
• Github: https://github.com/sfu-mial/IMAplusplus

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

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

#AI #DataScience #MachineLearning #HuggingFace #Research

✨Toward Stable Semi-Supervised Remote Sensing Segmentation via Co-Guidance and Co-Fusion

📝 Summary:
A semi-supervised remote sensing image segmentation framework combines vision-language and self-supervised models to reduce pseudo-label drift through dual-student architecture and semantic co-guidanc...

🔹 Publication Date: Published on Dec 28, 2025

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2512.23035
• PDF: https://arxiv.org/pdf/2512.23035
• Project Page: https://xavierjiezou.github.io/Co2S/
• Github: https://github.com/XavierJiezou/Co2S

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

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

#AI #DataScience #MachineLearning #HuggingFace #Research

✨Recursive Language Models

📝 Summary:
Recursive Language Models RLMs allow LLMs to process arbitrarily long prompts. RLMs programmatically decompose prompts and recursively call the LLM over snippets. This extends input length 100x and improves performance, even for shorter prompts, at similar cost.

🔹 Publication Date: Published on Dec 31, 2025

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2512.24601
• PDF: https://arxiv.org/pdf/2512.24601
• Github: https://github.com/alexzhang13/rlm/tree/main

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

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

#LLMs #AI #NLP #RecursiveLMs #LongContext

✨InfiniteVGGT: Visual Geometry Grounded Transformer for Endless Streams

📝 Summary:
InfiniteVGGT enables continuous 3D visual geometry understanding for infinite streams. It uses a causal transformer with adaptive rolling memory for long-term stability, outperforming existing streaming methods. A new Long3D benchmark is introduced for rigorous evaluation of such systems.

🔹 Publication Date: Published on Jan 5

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2601.02281
• PDF: https://arxiv.org/pdf/2601.02281
• Github: https://github.com/AutoLab-SAI-SJTU/InfiniteVGGT

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

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

#VisualGeometry #3DVision #Transformers #StreamingAI #DeepLearning

✨SWE-Lego: Pushing the Limits of Supervised Fine-tuning for Software Issue Resolving

📝 Summary:
SWE-Lego achieves state-of-the-art software issue resolution through a lightweight supervised fine-tuning approach. It uses a high-quality dataset and refined training procedures like error masking and a difficulty-based curriculum, outperforming complex methods. Performance is further boosted by...

🔹 Publication Date: Published on Jan 4

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2601.01426
• PDF: https://arxiv.org/pdf/2601.01426
• Project Page: https://github.com/SWE-Lego/SWE-Lego
• Github: https://github.com/SWE-Lego/SWE-Lego

🔹 Models citing this paper:
• https://huggingface.co/SWE-Lego/SWE-Lego-Qwen3-8B
• https://huggingface.co/SWE-Lego/SWE-Lego-Qwen3-32B

✨ Datasets citing this paper:
• https://huggingface.co/datasets/SWE-Lego/SWE-Lego-Real-Data
• https://huggingface.co/datasets/SWE-Lego/SWE-Lego-Synthetic-Data

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

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

#SoftwareEngineering #MachineLearning #LLM #FineTuning #AIforCode

✨M-ErasureBench: A Comprehensive Multimodal Evaluation Benchmark for Concept Erasure in Diffusion Models

📝 Summary:
Existing concept erasure methods in diffusion models are vulnerable to non-text inputs. M-ErasureBench is a new multimodal evaluation framework, and IRECE is a module to restore robustness against these attacks, reducing concept reproduction.

🔹 Publication Date: Published on Dec 28, 2025

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2512.22877
• PDF: https://arxiv.org/pdf/2512.22877

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

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

#DiffusionModels #ConceptErasure #MultimodalAI #AISafety #MachineLearning

I recommend you to use this app for people who want to make software or apps faster

Just send an prompt you want and he will build the program for you

✨Selective Imperfection as a Generative Framework for Analysis, Creativity and Discovery

📝 Summary:
Materiomusic links matter's hierarchical structures to music's compositional logic through vibrational principles. Sound serves as a scientific probe, revealing how selective imperfection drives novelty in both. AI models can leverage this framework for creative invention beyond interpolation.

🔹 Publication Date: Published on Dec 30, 2025

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2601.00863
• PDF: https://arxiv.org/pdf/2601.00863
• Github: https://github.com/lamm-mit/MusicAnalysis

✨ Datasets citing this paper:
• https://huggingface.co/datasets/lamm-mit/scales-12tet-defects

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

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

#GenerativeAI #ComputationalMusic #ComplexSystems #Creativity #Interdisciplinary

✨Confidence Estimation for LLMs in Multi-turn Interactions

📝 Summary:
This paper presents the first systematic study of confidence estimation in multi-turn LLM interactions, introducing a formal evaluation framework, novel metrics, and a Hinter-Guesser dataset paradigm. It reveals that current confidence techniques struggle with calibration and monotonicity in mult...

🔹 Publication Date: Published on Jan 5

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2601.02179
• PDF: https://arxiv.org/pdf/2601.02179

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

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

#LLM #ConfidenceEstimation #ConversationalAI #NLP #AIResearch

✨DiffProxy: Multi-View Human Mesh Recovery via Diffusion-Generated Dense Proxies

📝 Summary:
DiffProxy generates multi-view consistent human proxies using diffusion models to improve human mesh recovery. This bridges synthetic training and real-world generalization, achieving state-of-the-art performance on real benchmarks.

🔹 Publication Date: Published on Jan 5

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2601.02267
• PDF: https://arxiv.org/pdf/2601.02267
• Project Page: https://wrk226.github.io/DiffProxy.html
• Github: https://github.com/wrk226/DiffProxy

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

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

#HumanMeshRecovery #DiffusionModels #ComputerVision #DeepLearning #AI

✨CPPO: Contrastive Perception for Vision Language Policy Optimization

📝 Summary:
CPPO improves vision-language model fine-tuning by detecting perception tokens through entropy shifts. It then applies a Contrastive Perception Loss to enhance multimodal reasoning, outperforming prior methods more efficiently.

🔹 Publication Date: Published on Jan 1

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2601.00501
• PDF: https://arxiv.org/pdf/2601.00501

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

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

#VisionLanguageModels #MultimodalAI #ContrastiveLearning #DeepLearning #AIResearch

✨Prithvi-Complimentary Adaptive Fusion Encoder (CAFE): unlocking full-potential for flood inundation mapping

📝 Summary:
Prithvi-CAFE improves flood mapping by integrating a pretrained Geo-Foundation Model encoder with a parallel CNN branch featuring attention modules. This hybrid approach effectively captures both global context and critical local details, achieving state-of-the-art results on Sen1Flood11 and Floo...

🔹 Publication Date: Published on Jan 5

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2601.02315
• PDF: https://arxiv.org/pdf/2601.02315
• Github: https://github.com/Sk-2103/Prithvi-CAFE

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

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

#FloodMapping #DeepLearning #GeoAI #RemoteSensing #ComputerVision

nature papers: 1400$

Q1 and  Q2 papers    900$

Q3 and Q4 papers   500$

Doctoral thesis (complete)    700$

M.S thesis         300$

paper simulation   200$

Contact me
https://t.me/m/-nTmpj5vYzNk

✨LTX-2: Efficient Joint Audio-Visual Foundation Model

📝 Summary:
LTX-2 is an open-source audiovisual diffusion model generating synchronized video and audio content. It uses a dual-stream transformer to achieve state-of-the-art quality, producing rich audio tracks efficiently.

🔹 Publication Date: Published on Jan 6

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2601.03233
• PDF: https://arxiv.org/pdf/2601.03233

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

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

#AudiovisualAI #DiffusionModels #GenerativeAI #FoundationModels #VideoGeneration

✨InfiniDepth: Arbitrary-Resolution and Fine-Grained Depth Estimation with Neural Implicit Fields

📝 Summary:
InfiniDepth represents depth as neural implicit fields using a local implicit decoder, enabling continuous 2D coordinate querying for arbitrary-resolution depth estimation and superior performance in ...

🔹 Publication Date: Published on Jan 6

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2601.03252
• PDF: https://arxiv.org/pdf/2601.03252
• Github: https://zju3dv.github.io/InfiniDepth

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

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

#AI #DataScience #MachineLearning #HuggingFace #Research

✨FFP-300K: Scaling First-Frame Propagation for Generalizable Video Editing

📝 Summary:
A new large-scale video dataset and framework are presented that enable effective first-frame propagation without runtime guidance through adaptive spatio-temporal positional encoding and self-distill...

🔹 Publication Date: Published on Jan 5

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2601.01720
• PDF: https://arxiv.org/pdf/2601.01720
• Project Page: https://ffp-300k.github.io/

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

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

#AI #DataScience #MachineLearning #HuggingFace #Research

✨MOSS Transcribe Diarize: Accurate Transcription with Speaker Diarization

📝 Summary:
A unified multimodal large language model for end-to-end speaker-attributed, time-stamped transcription with extended context window and strong generalization across benchmarks. AI-generated summary S...

🔹 Publication Date: Published on Jan 4

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2601.01554
• PDF: https://arxiv.org/pdf/2601.01554
• Project Page: https://mosi.cn/models/moss-transcribe-diarize

✨ Spaces citing this paper:
• https://huggingface.co/spaces/OpenMOSS-Team/MOSS-transcribe-diarize

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

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

#AI #DataScience #MachineLearning #HuggingFace #Research