✨Guidelines to Prompt Large Language Models for Code Generation: An Empirical Characterization
📝 Summary:
Research derives and evaluates prompt optimization guidelines for code generation tasks in software engineering, identifying 10 specific improvement patterns related to input/output specification, con...
🔹 Publication Date: Published on Jan 19
🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2601.13118
• PDF: https://arxiv.org/pdf/2601.13118
==================================
For more data science resources:
✓ https://news.1rj.ru/str/DataScienceT
#AI #DataScience #MachineLearning #HuggingFace #Research
📝 Summary:
Research derives and evaluates prompt optimization guidelines for code generation tasks in software engineering, identifying 10 specific improvement patterns related to input/output specification, con...
🔹 Publication Date: Published on Jan 19
🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2601.13118
• PDF: https://arxiv.org/pdf/2601.13118
==================================
For more data science resources:
✓ https://news.1rj.ru/str/DataScienceT
#AI #DataScience #MachineLearning #HuggingFace #Research
❤1👍1
✨Knowledge is Not Enough: Injecting RL Skills for Continual Adaptation
📝 Summary:
LLMs struggle to apply new knowledge effectively via SFT alone. PaST combines SFT with injecting a domain-agnostic Skill Vector, derived from RL, to efficiently transfer reasoning skills. This novel framework significantly improves performance in question answering and tool-use tasks.
🔹 Publication Date: Published on Jan 16
🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2601.11258
• PDF: https://arxiv.org/pdf/2601.11258
==================================
For more data science resources:
✓ https://news.1rj.ru/str/DataScienceT
#LLM #ReinforcementLearning #ContinualLearning #AI #MachineLearning
📝 Summary:
LLMs struggle to apply new knowledge effectively via SFT alone. PaST combines SFT with injecting a domain-agnostic Skill Vector, derived from RL, to efficiently transfer reasoning skills. This novel framework significantly improves performance in question answering and tool-use tasks.
🔹 Publication Date: Published on Jan 16
🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2601.11258
• PDF: https://arxiv.org/pdf/2601.11258
==================================
For more data science resources:
✓ https://news.1rj.ru/str/DataScienceT
#LLM #ReinforcementLearning #ContinualLearning #AI #MachineLearning
❤1
✨Dancing in Chains: Strategic Persuasion in Academic Rebuttal via Theory of Mind
📝 Summary:
RebuttalAgent is a novel AI framework that applies Theory of Mind to academic rebuttal. It models reviewer mental states to formulate strategic, persuasive responses, significantly outperforming existing models.
🔹 Publication Date: Published on Jan 22
🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2601.15715
• PDF: https://arxiv.org/pdf/2601.15715
==================================
For more data science resources:
✓ https://news.1rj.ru/str/DataScienceT
#AI #TheoryOfMind #AcademicRebuttal #NLP #MachineLearning
📝 Summary:
RebuttalAgent is a novel AI framework that applies Theory of Mind to academic rebuttal. It models reviewer mental states to formulate strategic, persuasive responses, significantly outperforming existing models.
🔹 Publication Date: Published on Jan 22
🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2601.15715
• PDF: https://arxiv.org/pdf/2601.15715
==================================
For more data science resources:
✓ https://news.1rj.ru/str/DataScienceT
#AI #TheoryOfMind #AcademicRebuttal #NLP #MachineLearning
❤2
✨GameTalk: Training LLMs for Strategic Conversation
📝 Summary:
The GameTalk framework trains large language models for strategic multi-turn dialogue, optimizing global objectives using whole-conversation reward signals. This approach significantly outperforms untrained models, showing conversational fine-tuning is a promising path for LLM reasoning and negot...
🔹 Publication Date: Published on Jan 22
🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2601.16276
• PDF: https://arxiv.org/pdf/2601.16276
==================================
For more data science resources:
✓ https://news.1rj.ru/str/DataScienceT
#LLMs #ConversationalAI #StrategicDialogue #AITraining #AIReasoning
📝 Summary:
The GameTalk framework trains large language models for strategic multi-turn dialogue, optimizing global objectives using whole-conversation reward signals. This approach significantly outperforms untrained models, showing conversational fine-tuning is a promising path for LLM reasoning and negot...
🔹 Publication Date: Published on Jan 22
🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2601.16276
• PDF: https://arxiv.org/pdf/2601.16276
==================================
For more data science resources:
✓ https://news.1rj.ru/str/DataScienceT
#LLMs #ConversationalAI #StrategicDialogue #AITraining #AIReasoning
❤1
✨ChartVerse: Scaling Chart Reasoning via Reliable Programmatic Synthesis from Scratch
📝 Summary:
ChartVerse is a framework that synthesizes complex charts and reliable reasoning data for VLMs. It uses a novel metric, Rollout Posterior Entropy, for complexity-aware chart generation and an answer-first QA synthesis to ensure reasoning rigor. This leads to state-of-the-art performance in chart ...
🔹 Publication Date: Published on Jan 20
🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2601.13606
• PDF: https://arxiv.org/pdf/2601.13606
• Project Page: https://chartverse.github.io/
• Github: https://github.com/starriver030515/ChartVerse
🔹 Models citing this paper:
• https://huggingface.co/opendatalab/ChartVerse-Coder
• https://huggingface.co/opendatalab/ChartVerse-2B
• https://huggingface.co/opendatalab/ChartVerse-8B
✨ Datasets citing this paper:
• https://huggingface.co/datasets/opendatalab/ChartVerse-SFT-1800K
• https://huggingface.co/datasets/opendatalab/ChartVerse-SFT-600K
• https://huggingface.co/datasets/opendatalab/ChartVerse-RL-40K
==================================
For more data science resources:
✓ https://news.1rj.ru/str/DataScienceT
#AI #VLMs #ChartReasoning #MachineLearning #DataScience
📝 Summary:
ChartVerse is a framework that synthesizes complex charts and reliable reasoning data for VLMs. It uses a novel metric, Rollout Posterior Entropy, for complexity-aware chart generation and an answer-first QA synthesis to ensure reasoning rigor. This leads to state-of-the-art performance in chart ...
🔹 Publication Date: Published on Jan 20
🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2601.13606
• PDF: https://arxiv.org/pdf/2601.13606
• Project Page: https://chartverse.github.io/
• Github: https://github.com/starriver030515/ChartVerse
🔹 Models citing this paper:
• https://huggingface.co/opendatalab/ChartVerse-Coder
• https://huggingface.co/opendatalab/ChartVerse-2B
• https://huggingface.co/opendatalab/ChartVerse-8B
✨ Datasets citing this paper:
• https://huggingface.co/datasets/opendatalab/ChartVerse-SFT-1800K
• https://huggingface.co/datasets/opendatalab/ChartVerse-SFT-600K
• https://huggingface.co/datasets/opendatalab/ChartVerse-RL-40K
==================================
For more data science resources:
✓ https://news.1rj.ru/str/DataScienceT
#AI #VLMs #ChartReasoning #MachineLearning #DataScience
arXiv.org
ChartVerse: Scaling Chart Reasoning via Reliable Programmatic...
Chart reasoning is a critical capability for Vision Language Models (VLMs). However, the development of open-source models is severely hindered by the lack of high-quality training data. Existing...
❤1
Forwarded from Machine Learning with Python
Do you see yourself as a programmer, researcher, or engineer?
Anonymous Poll
45%
Programmer
22%
Researcher
33%
Engineer
✨VISTA-PATH: An interactive foundation model for pathology image segmentation and quantitative analysis in computational pathology
📝 Summary:
VISTA-PATH is an interactive, class-aware foundation model for pathology image segmentation. It integrates visual context, semantic denoscriptions, and expert feedback for precise multi-class segmentation, outperforming existing models. This high-fidelity segmentation supports clinical interpretati...
🔹 Publication Date: Published on Jan 23
🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2601.16451
• PDF: https://arxiv.org/pdf/2601.16451
==================================
For more data science resources:
✓ https://news.1rj.ru/str/DataScienceT
#ComputationalPathology #AIinMedicine #MedicalImaging #FoundationModels #PathologyAI
📝 Summary:
VISTA-PATH is an interactive, class-aware foundation model for pathology image segmentation. It integrates visual context, semantic denoscriptions, and expert feedback for precise multi-class segmentation, outperforming existing models. This high-fidelity segmentation supports clinical interpretati...
🔹 Publication Date: Published on Jan 23
🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2601.16451
• PDF: https://arxiv.org/pdf/2601.16451
==================================
For more data science resources:
✓ https://news.1rj.ru/str/DataScienceT
#ComputationalPathology #AIinMedicine #MedicalImaging #FoundationModels #PathologyAI
✨IVRA: Improving Visual-Token Relations for Robot Action Policy with Training-Free Hint-Based Guidance
📝 Summary:
IVRA improves spatial understanding in VLA models by training-free injection of vision encoder affinity signals into language model layers at inference time. This enhances geometric structure and robot action policies. It shows consistent performance gains across diverse 2D and 3D manipulation ta...
🔹 Publication Date: Published on Jan 22
🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2601.16207
• PDF: https://arxiv.org/pdf/2601.16207
• Github: https://jongwoopark7978.github.io/IVRA
==================================
For more data science resources:
✓ https://news.1rj.ru/str/DataScienceT
#Robotics #VisionLanguageModels #SpatialAI #RobotLearning #DeepLearning
📝 Summary:
IVRA improves spatial understanding in VLA models by training-free injection of vision encoder affinity signals into language model layers at inference time. This enhances geometric structure and robot action policies. It shows consistent performance gains across diverse 2D and 3D manipulation ta...
🔹 Publication Date: Published on Jan 22
🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2601.16207
• PDF: https://arxiv.org/pdf/2601.16207
• Github: https://jongwoopark7978.github.io/IVRA
==================================
For more data science resources:
✓ https://news.1rj.ru/str/DataScienceT
#Robotics #VisionLanguageModels #SpatialAI #RobotLearning #DeepLearning
✨Prometheus: Unified Knowledge Graphs for Issue Resolution in Multilingual Codebases
📝 Summary:
Prometheus is a multi-agent system that uses a unified knowledge graph of code repositories to resolve real-world issues across multiple programming languages. It improves upon existing methods by handling diverse languages and real-world scenarios.
🔹 Publication Date: Published on Jul 26, 2025
🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2507.19942
• PDF: https://arxiv.org/pdf/2507.19942
• Github: https://github.com/Pantheon-temple/Prometheus
==================================
For more data science resources:
✓ https://news.1rj.ru/str/DataScienceT
#KnowledgeGraphs #MultiAgentSystems #CodeAnalysis #SoftwareEngineering #AI
📝 Summary:
Prometheus is a multi-agent system that uses a unified knowledge graph of code repositories to resolve real-world issues across multiple programming languages. It improves upon existing methods by handling diverse languages and real-world scenarios.
🔹 Publication Date: Published on Jul 26, 2025
🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2507.19942
• PDF: https://arxiv.org/pdf/2507.19942
• Github: https://github.com/Pantheon-temple/Prometheus
==================================
For more data science resources:
✓ https://news.1rj.ru/str/DataScienceT
#KnowledgeGraphs #MultiAgentSystems #CodeAnalysis #SoftwareEngineering #AI
✨The Script is All You Need: An Agentic Framework for Long-Horizon Dialogue-to-Cinematic Video Generation
📝 Summary:
This paper presents an agentic framework translating dialogue into cinematic videos. ScripterAgent generates a noscript from dialogue, which DirectorAgent uses to orchestrate video models for long-horizon coherence. The system improves noscript faithfulness and reveals a trade-off in current video ge...
🔹 Publication Date: Published on Jan 25
🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2601.17737
• PDF: https://arxiv.org/pdf/2601.17737
• Project Page: https://xd-mu.github.io/ScriptIsAllYouNeed/
• Github: https://github.com/Tencent/digitalhuman/tree/main/ScriptAgent
==================================
For more data science resources:
✓ https://news.1rj.ru/str/DataScienceT
#AIAgents #VideoGeneration #GenerativeAI #MultimodalAI #DeepLearning
📝 Summary:
This paper presents an agentic framework translating dialogue into cinematic videos. ScripterAgent generates a noscript from dialogue, which DirectorAgent uses to orchestrate video models for long-horizon coherence. The system improves noscript faithfulness and reveals a trade-off in current video ge...
🔹 Publication Date: Published on Jan 25
🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2601.17737
• PDF: https://arxiv.org/pdf/2601.17737
• Project Page: https://xd-mu.github.io/ScriptIsAllYouNeed/
• Github: https://github.com/Tencent/digitalhuman/tree/main/ScriptAgent
==================================
For more data science resources:
✓ https://news.1rj.ru/str/DataScienceT
#AIAgents #VideoGeneration #GenerativeAI #MultimodalAI #DeepLearning
✨Elastic Attention: Test-time Adaptive Sparsity Ratios for Efficient Transformers
📝 Summary:
Elastic Attention dynamically adjusts transformer sparsity ratios during inference using a lightweight Attention Router. This resolves static sparsity limitations in existing models, boosting efficiency and performance for long-context LLMs with minimal training.
🔹 Publication Date: Published on Jan 24
🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2601.17367
• PDF: https://arxiv.org/pdf/2601.17367
• Project Page: https://github.com/LCM-Lab/Elastic-Attention
• Github: https://github.com/LCM-Lab/Elastic-Attention
==================================
For more data science resources:
✓ https://news.1rj.ru/str/DataScienceT
#Transformers #LLMs #Sparsity #DeepLearning #EfficientAI
📝 Summary:
Elastic Attention dynamically adjusts transformer sparsity ratios during inference using a lightweight Attention Router. This resolves static sparsity limitations in existing models, boosting efficiency and performance for long-context LLMs with minimal training.
🔹 Publication Date: Published on Jan 24
🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2601.17367
• PDF: https://arxiv.org/pdf/2601.17367
• Project Page: https://github.com/LCM-Lab/Elastic-Attention
• Github: https://github.com/LCM-Lab/Elastic-Attention
==================================
For more data science resources:
✓ https://news.1rj.ru/str/DataScienceT
#Transformers #LLMs #Sparsity #DeepLearning #EfficientAI
✨Can LLMs Clean Up Your Mess? A Survey of Application-Ready Data Preparation with LLMs
📝 Summary:
This paper surveys how LLMs are transforming data preparation tasks like cleaning, integration, and enrichment. It details the shift from rule-based to prompt-driven approaches, outlining techniques, benefits, and challenges, along with future research directions.
🔹 Publication Date: Published on Jan 22
🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2601.17058
• PDF: https://arxiv.org/pdf/2601.17058
• Project Page: https://github.com/weAIDB/awesome-data-llm
• Github: https://github.com/weAIDB/awesome-data-llm
==================================
For more data science resources:
✓ https://news.1rj.ru/str/DataScienceT
#LLMs #DataPreparation #DataCleaning #DataScience #AI
📝 Summary:
This paper surveys how LLMs are transforming data preparation tasks like cleaning, integration, and enrichment. It details the shift from rule-based to prompt-driven approaches, outlining techniques, benefits, and challenges, along with future research directions.
🔹 Publication Date: Published on Jan 22
🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2601.17058
• PDF: https://arxiv.org/pdf/2601.17058
• Project Page: https://github.com/weAIDB/awesome-data-llm
• Github: https://github.com/weAIDB/awesome-data-llm
==================================
For more data science resources:
✓ https://news.1rj.ru/str/DataScienceT
#LLMs #DataPreparation #DataCleaning #DataScience #AI
✨VIBEVOICE-ASR Technical Report
📝 Summary:
VibeVoice-ASR is a unified end-to-end speech understanding framework that processes long-form audio in a single pass while supporting multilingual, code-switching, and domain-specific context injectio...
🔹 Publication Date: Published on Jan 26
🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2601.18184
• PDF: https://arxiv.org/pdf/2601.18184
==================================
For more data science resources:
✓ https://news.1rj.ru/str/DataScienceT
#AI #DataScience #MachineLearning #HuggingFace #Research
📝 Summary:
VibeVoice-ASR is a unified end-to-end speech understanding framework that processes long-form audio in a single pass while supporting multilingual, code-switching, and domain-specific context injectio...
🔹 Publication Date: Published on Jan 26
🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2601.18184
• PDF: https://arxiv.org/pdf/2601.18184
==================================
For more data science resources:
✓ https://news.1rj.ru/str/DataScienceT
#AI #DataScience #MachineLearning #HuggingFace #Research
✨Scientific Image Synthesis: Benchmarking, Methodologies, and Downstream Utility
📝 Summary:
Scientific image synthesis using logic-driven frameworks like ImgCoder improves multimodal reasoning by addressing visual-logic divergence through structured generation and evaluation benchmarks. AI-g...
🔹 Publication Date: Published on Jan 17
🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2601.17027
• PDF: https://arxiv.org/pdf/2601.17027
• Project Page: https://scigenbench.github.io/
• Github: https://github.com/SciGenBench/SciGenBench
==================================
For more data science resources:
✓ https://news.1rj.ru/str/DataScienceT
#AI #DataScience #MachineLearning #HuggingFace #Research
📝 Summary:
Scientific image synthesis using logic-driven frameworks like ImgCoder improves multimodal reasoning by addressing visual-logic divergence through structured generation and evaluation benchmarks. AI-g...
🔹 Publication Date: Published on Jan 17
🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2601.17027
• PDF: https://arxiv.org/pdf/2601.17027
• Project Page: https://scigenbench.github.io/
• Github: https://github.com/SciGenBench/SciGenBench
==================================
For more data science resources:
✓ https://news.1rj.ru/str/DataScienceT
#AI #DataScience #MachineLearning #HuggingFace #Research
✨AR-Omni: A Unified Autoregressive Model for Any-to-Any Generation
📝 Summary:
AR-Omni is a unified autoregressive model for any-to-any multimodal generation using a single Transformer. It generates text images and streaming speech without relying on expert components. The model addresses key challenges like modality imbalance and achieves strong real-time quality.
🔹 Publication Date: Published on Jan 25
🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2601.17761
• PDF: https://arxiv.org/pdf/2601.17761
• Project Page: https://modalitydance.github.io/AR-Omni
• Github: https://modalitydance.github.io/AR-Omni
🔹 Models citing this paper:
• https://huggingface.co/ModalityDance/AR-Omni-Pretrain-v0.1
• https://huggingface.co/ModalityDance/AR-Omni-Chat-v0.1
✨ Datasets citing this paper:
• https://huggingface.co/datasets/ModalityDance/AR-Omni-Instruct-v0.1
==================================
For more data science resources:
✓ https://news.1rj.ru/str/DataScienceT
#AI #DataScience #MachineLearning #HuggingFace #Research
📝 Summary:
AR-Omni is a unified autoregressive model for any-to-any multimodal generation using a single Transformer. It generates text images and streaming speech without relying on expert components. The model addresses key challenges like modality imbalance and achieves strong real-time quality.
🔹 Publication Date: Published on Jan 25
🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2601.17761
• PDF: https://arxiv.org/pdf/2601.17761
• Project Page: https://modalitydance.github.io/AR-Omni
• Github: https://modalitydance.github.io/AR-Omni
🔹 Models citing this paper:
• https://huggingface.co/ModalityDance/AR-Omni-Pretrain-v0.1
• https://huggingface.co/ModalityDance/AR-Omni-Chat-v0.1
✨ Datasets citing this paper:
• https://huggingface.co/datasets/ModalityDance/AR-Omni-Instruct-v0.1
==================================
For more data science resources:
✓ https://news.1rj.ru/str/DataScienceT
#AI #DataScience #MachineLearning #HuggingFace #Research
arXiv.org
AR-Omni: A Unified Autoregressive Model for Any-to-Any Generation
Real-world perception and interaction are inherently multimodal, encompassing not only language but also vision and speech, which motivates the development of "Omni" MLLMs that support both...
✨Least-Loaded Expert Parallelism: Load Balancing An Imbalanced Mixture-of-Experts
📝 Summary:
Imbalanced expert routing in Mixture-of-Experts models leads to computational inefficiencies in expert parallelism, which are addressed by a dynamic rerouting algorithm that balances workload and redu...
🔹 Publication Date: Published on Jan 23
🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2601.17111
• PDF: https://arxiv.org/pdf/2601.17111
==================================
For more data science resources:
✓ https://news.1rj.ru/str/DataScienceT
#AI #DataScience #MachineLearning #HuggingFace #Research
📝 Summary:
Imbalanced expert routing in Mixture-of-Experts models leads to computational inefficiencies in expert parallelism, which are addressed by a dynamic rerouting algorithm that balances workload and redu...
🔹 Publication Date: Published on Jan 23
🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2601.17111
• PDF: https://arxiv.org/pdf/2601.17111
==================================
For more data science resources:
✓ https://news.1rj.ru/str/DataScienceT
#AI #DataScience #MachineLearning #HuggingFace #Research
✨DRPG (Decompose, Retrieve, Plan, Generate): An Agentic Framework for Academic Rebuttal
📝 Summary:
An agentic framework for automatic academic rebuttal generation that decomposes reviews, retrieves evidence, plans rebuttal strategies, and generates persuasive responses with human-level performance ...
🔹 Publication Date: Published on Jan 26
🔹 Paper Links:
• arXiv Page: https://huggingface.co/collections/HakHan/drpg-rebuttalagent
• PDF: https://arxiv.org/pdf/2601.18081
• Github: https://github.com/ulab-uiuc/DRPG-RebuttalAgent/tree/master
==================================
For more data science resources:
✓ https://news.1rj.ru/str/DataScienceT
#AI #DataScience #MachineLearning #HuggingFace #Research
📝 Summary:
An agentic framework for automatic academic rebuttal generation that decomposes reviews, retrieves evidence, plans rebuttal strategies, and generates persuasive responses with human-level performance ...
🔹 Publication Date: Published on Jan 26
🔹 Paper Links:
• arXiv Page: https://huggingface.co/collections/HakHan/drpg-rebuttalagent
• PDF: https://arxiv.org/pdf/2601.18081
• Github: https://github.com/ulab-uiuc/DRPG-RebuttalAgent/tree/master
==================================
For more data science resources:
✓ https://news.1rj.ru/str/DataScienceT
#AI #DataScience #MachineLearning #HuggingFace #Research
❤1
✨iFSQ: Improving FSQ for Image Generation with 1 Line of Code
📝 Summary:
Finite Scalar Quantization with improved activation mapping enables unified modeling of discrete and continuous image generation approaches, revealing optimal representation balance and performance ch...
🔹 Publication Date: Published on Jan 23
🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2601.17124
• PDF: https://arxiv.org/pdf/2601.17124
• Github: https://github.com/Tencent-Hunyuan/iFSQ
==================================
For more data science resources:
✓ https://news.1rj.ru/str/DataScienceT
#AI #DataScience #MachineLearning #HuggingFace #Research
📝 Summary:
Finite Scalar Quantization with improved activation mapping enables unified modeling of discrete and continuous image generation approaches, revealing optimal representation balance and performance ch...
🔹 Publication Date: Published on Jan 23
🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2601.17124
• PDF: https://arxiv.org/pdf/2601.17124
• Github: https://github.com/Tencent-Hunyuan/iFSQ
==================================
For more data science resources:
✓ https://news.1rj.ru/str/DataScienceT
#AI #DataScience #MachineLearning #HuggingFace #Research
This media is not supported in your browser
VIEW IN TELEGRAM
✨Self-Refining Video Sampling
📝 Summary:
Self-refining video sampling improves motion coherence and physics alignment by using a pre-trained video generator as its own denoising autoencoder for iterative refinement with uncertainty-aware reg...
🔹 Publication Date: Published on Jan 26
🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2601.18577
• PDF: https://arxiv.org/pdf/2601.18577
• Project Page: https://agwmon.github.io/self-refine-video/
• Github: https://github.com/agwmon/self-refine-video
==================================
For more data science resources:
✓ https://news.1rj.ru/str/DataScienceT
#AI #DataScience #MachineLearning #HuggingFace #Research
📝 Summary:
Self-refining video sampling improves motion coherence and physics alignment by using a pre-trained video generator as its own denoising autoencoder for iterative refinement with uncertainty-aware reg...
🔹 Publication Date: Published on Jan 26
🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2601.18577
• PDF: https://arxiv.org/pdf/2601.18577
• Project Page: https://agwmon.github.io/self-refine-video/
• Github: https://github.com/agwmon/self-refine-video
==================================
For more data science resources:
✓ https://news.1rj.ru/str/DataScienceT
#AI #DataScience #MachineLearning #HuggingFace #Research