🔥 Adam's Law: Textual Frequency Law on Large Language Models

💡 The paper proposes a novel framework to improve large language model performance through textual frequency analysis. The authors argue that textual frequency, which is the frequency of certain words or phrases in a language, is relevant to human cognition and can also be applied to large language models. However, this topic has been understudied in the context of large language models.

The proposed framework consists of three main components. First, the authors introduce the Textual Frequency Law, which states that frequent textual data should be preferred for large language models, both for prompting and fine-tuning. To estimate the sentence-level frequency, the authors use online resources, as many large language models are closed-source in their training data. They also utilize an input paraphraser to paraphrase the input into a more frequent textual expression.

The second component is Textual Frequency Distillation, which involves querying large language models to conduct story completion by extending sentences in the datasets. The resulting corpora are used to adjust the initial estimation of textual frequency.

The third component is Curriculum Textual Frequency Training, which fine-tunes large language models in an increasing order of sentence-level frequency. This means that the models are first trained on the most frequent sentences and then gradually moved to less frequent ones.

The authors conducted experiments on a curated dataset called Textual Frequency Paired Dataset, which covers tasks such as math reasoning, machine translation, commonsense reasoning, and agentic tool calling. The results show that the proposed framework is effective in improving large language model performance.

Overall, the paper contributes to the understanding of textual frequency in large language models and provides a novel framework for improving their performance. The proposed framework has the potential to be applied to various natural language processing tasks and can lead to more efficient and effective large language models.

📅 Published on Apr 2

🔗 Links:
• arXiv: https://arxiv.org/abs/2604.02176
• PDF: https://arxiv.org/pdf/2604.02176
• GitHub: https://github.com/HongyuanLuke/frequencylaw ⭐ 658

📊 Datasets citing this paper:
• https://huggingface.co/datasets/Akaashiiii/TFPD

━━━━━━━━━━━━━━━━━━━━━━━━
📢 By: https://xn--r1a.website/PaperNexus

#AdamSLaw #TextualFrequencyAnalysis #LargeLanguageModels #NaturalLanguageProcessing #LanguageModelOptimization

Unlock Your AI Career
Join our Data Science Full Stack with AI Course – a real-time, project-based online training designed for hands-on mastery.
Core Topics Covered
•  Data Science using Python with Generative AI: Build end-to-end data pipelines, from data wrangling to deploying AI models with Python libraries like Pandas, Scikit-learn, and Hugging Face transformers.
•  Prompt Engineering: Craft precise prompts to maximize output from models like GPT and Gemini for accurate, creative results.
•  AI Agents & Agentic AI: Develop autonomous agents that reason, plan, and act using frameworks like Lang Chain for real-world automation.
Why Choose This Course?
This training emphasizes live sessions, industry projects, and practical skills for immediate job impact, similar to top programs offering 100+ hours of Python-to-AI progression.
Ready to start? Call/WhatsApp: (+91)-7416877757
WhatsApp Link:-
http://wa.me/+917416877757

🔥 RLDX-1 Technical Report

💡 The paper introduces RLDX-1, a general-purpose robotic policy for dexterous manipulation that addresses the limitations of existing vision-language-action models. These models have shown progress in human-like generalist robotic policies but struggle with complex real-world tasks that require broader functional capabilities such as motion awareness, memory-aware decision making, and physical sensing. To overcome this, RLDX-1 uses a Multi-Stream Action Transformer architecture that integrates heterogeneous modalities through modality-specific streams with cross-modal joint self-attention. This architecture is combined with system-level design choices including synthesizing training data for rare manipulation scenarios, learning procedures specialized for human-like manipulation, and inference optimizations for real-time deployment. The results show that RLDX-1 outperforms recent frontier vision-language-action models across both simulation benchmarks and real-world tasks, achieving success rates of 86.8 percent in ALLEX humanoid tasks compared to around 40 percent for other models. This positions RLDX-1 as a promising step toward reliable vision-language-action models for complex and dynamic real-world dexterous manipulation. The method and results demonstrate the ability of RLDX-1 to control a high-degree-of-freedom humanoid robot under diverse functional demands, highlighting its potential for complex real-world tasks.

📅 Published on May 5

🔗 Links:
• arXiv: https://arxiv.org/abs/2605.03269
• PDF: https://arxiv.org/pdf/2605.03269
• Project Page: http://rlwrld.ai/rldx-1
• GitHub: https://github.com/RLWRLD/RLDX-1 ⭐ 75

🤖 Models citing this paper:
• https://huggingface.co/RLWRLD/RLDX-1-PT
• https://huggingface.co/RLWRLD/RLDX-1-FT-ROBOCASA
• https://huggingface.co/RLWRLD/RLDX-1-MT-ALLEX

━━━━━━━━━━━━━━━━━━━━━━━━
📢 By: https://xn--r1a.website/PaperNexus

#RoboticManipulation #DexterousRobotics #VisionLanguageAction #MultiModalLearning #RobotPolicyLearning

🔥 PhysForge: Generating Physics-Grounded 3D Assets for Interactive Virtual World

💡 The paper introduces PhysForge, a system for generating interactive 3D assets that combines visual-language modeling with a physics-grounded diffusion model. The problem addressed is the lack of functional properties in existing methods for generating 3D assets, which focus on static geometry and overlook the need for interactive virtual worlds and embodied AI. To solve this, PhysForge uses a two-stage framework, first using a visual-language model to plan a hierarchical physical blueprint that defines material, functional, and kinematic constraints. Then, a physics-grounded diffusion model synthesizes high-fidelity geometry and precise kinematic parameters using a novel injection mechanism called KineVoxel Injection. The system is supported by PhysDB, a large-scale dataset of 150,000 assets with physical annotations. The results show that PhysForge produces functionally plausible and simulation-ready assets, providing a robust data engine for interactive 3D content and embodied agents. Overall, PhysForge contributes a new approach to generating physics-grounded 3D assets that can be used in interactive virtual worlds and embodied AI applications.

📅 Published on May 6

🔗 Links:
• arXiv: https://arxiv.org/abs/2605.05163
• PDF: https://arxiv.org/pdf/2605.05163
• Project Page: https://hku-mmlab.github.io/PhysForge/
• GitHub: https://github.com/HKU-MMLab/PhysForge ⭐ 41

━━━━━━━━━━━━━━━━━━━━━━━━
📢 By: https://xn--r1a.website/PaperNexus

#PhysicsGroundedModeling #InteractiveVirtualWorlds #3DAssetGeneration #EmbodiedAI #PhysicsBasedRendering

🔥 D-OPSD: On-Policy Self-Distillation for Continuously Tuning Step-Distilled Diffusion Models

💡 The paper introduces D-OPSD, a new training approach for diffusion models that enables efficient supervised fine-tuning while preserving few-step inference capabilities. The current landscape of high-performance image generation models is shifting from inefficient multi-step models to efficient few-step models, but these models are challenging to fine-tune using traditional techniques. The problem with traditional fine-tuning methods is that they compromise the model's inherent few-step inference capability.

To address this issue, the authors propose D-OPSD, which leverages on-policy self-distillation with text and multimodal features. The method works by making the model act as both the teacher and the student, where the student is conditioned only on the text feature, and the teacher is conditioned on the multimodal feature of both the text prompt and the target image. The training process minimizes the difference between the predicted distributions over the student's own roll-outs, allowing the model to learn new concepts and styles without sacrificing its original few-step capacity.

The key contribution of D-OPSD is that it enables on-policy learning during supervised fine-tuning, which allows the model to learn from its own trajectory and under its own supervision. This approach enables the model to inherit the in-context capabilities of its encoder, making it possible to fine-tune the model continuously without compromising its few-step inference capability. The results show that D-OPSD enables efficient supervised fine-tuning for diffusion models, making it a promising approach for high-performance image generation models.

📅 Published on May 6

🔗 Links:
• arXiv: https://arxiv.org/abs/2605.05204
• PDF: https://arxiv.org/pdf/2605.05204
• Project Page: https://vvvvvjdy.github.io/d-opsd/
• GitHub: https://github.com/vvvvvjdy/D-OPSD ⭐ 24

━━━━━━━━━━━━━━━━━━━━━━━━
📢 By: https://xn--r1a.website/PaperNexus

#DiffusionModels #SelfDistillation #FewShotLearning #ImageGeneration #MultimodalLearning

🔥 MedSkillAudit: A Domain-Specific Audit Framework for Medical Research Agent Skills

💡 The paper presents a domain specific audit framework called MedSkillAudit for assessing the quality of medical research agent skills. The problem addressed is the need for reliable evaluation of these skills to ensure scientific integrity, methodological validity, reproducibility, and boundary safety in healthcare applications.

The authors developed MedSkillAudit, a layered framework that evaluates skill release readiness before deployment. They applied this framework to 75 skills across five medical research categories and compared the results with expert reviews. Two experts independently assigned quality scores and release dispositions to each skill, and the system expert agreement was quantified using statistical measures.

The results show that MedSkillAudit achieved a high level of agreement with expert reviews, with a mean consensus quality score of 72.4. The framework was able to identify skills that fell below the Limited Release threshold, with 57.3 percent of skills requiring further development. The system expert agreement was higher than the human inter-rater agreement, indicating that MedSkillAudit can provide a reliable and consistent evaluation of medical research agent skills.

The study found that MedSkillAudit performed well in certain categories, such as Protocol Design, but had a negative agreement in Academic Writing, suggesting a structural mismatch between the rubric and expert reviews. Overall, the paper concludes that domain specific pre-deployment audit frameworks like MedSkillAudit can provide a practical foundation for governing medical research agent skills, complementing general purpose quality checks with structured audit workflows tailored to scientific use cases.

The contributions of the paper are the development and evaluation of a domain specific audit framework for medical research agent skills, and the demonstration of its reliability and consistency in assessing the quality of these skills. The study provides a foundation for further research in this area and highlights the importance of domain specific evaluation frameworks for ensuring the quality and safety of AI systems in healthcare applications.

📅 Published on Apr 22

🔗 Links:
• arXiv: https://arxiv.org/abs/2604.20441
• PDF: https://arxiv.org/pdf/2604.20441
• GitHub: https://github.com/aipoch/medical-research-skills ⭐ 535

━━━━━━━━━━━━━━━━━━━━━━━━
📢 By: https://xn--r1a.website/PaperNexus

#MedicalResearchIntegrity #AgentSkillEvaluation #HealthcareAuditFrameworks #ScientificValidityAssessment #ReproducibilityInMedicine

🔥 Awaking Spatial Intelligence in Unified Multimodal Understanding and Generation

💡 The paper presents JoyAI-Image, a unified multimodal foundation model that integrates visual understanding, text-to-image generation, and instruction-guided image editing. The model combines a spatially enhanced Multimodal Large Language Model with a Multimodal Diffusion Transformer, allowing for a shared multimodal interface between perception and generation. The authors propose a scalable training recipe that incorporates unified instruction tuning, long-text rendering supervision, spatially grounded data, and general and spatial editing signals. This design enables the model to achieve broad multimodal capabilities while strengthening geometry-aware reasoning and controllable visual synthesis. The experiments demonstrate that JoyAI-Image achieves state-of-the-art or highly competitive performance across various benchmarks, including understanding, generation, long-text rendering, and editing tasks. The model's bidirectional loop between enhanced understanding, controllable spatial editing, and novel-view-assisted reasoning enables it to move beyond general visual competence toward stronger spatial intelligence. The results suggest a promising path for unified visual models in downstream applications such as vision-language-action systems and world models. Overall, the paper contributes to the development of a unified multimodal model that can effectively understand and generate visual content with enhanced spatial intelligence.

📅 Published on May 5

🔗 Links:
• arXiv: https://arxiv.org/abs/2605.04128
• PDF: https://arxiv.org/pdf/2605.04128
• GitHub: https://github.com/jd-opensource/JoyAI-Image ⭐ 2.1k

🤖 Models citing this paper:
• https://huggingface.co/jdopensource/JoyAI-Image-Edit

🚀 Spaces citing this paper:
• https://huggingface.co/spaces/stevengrove/JoyAI-Image-Edit-Space
• https://huggingface.co/spaces/stevengrove/JoyAI-Image-Edit
• https://huggingface.co/spaces/Merlinus001/JoyAI-Image-Edit-Space

━━━━━━━━━━━━━━━━━━━━━━━━
📢 By: https://xn--r1a.website/PaperNexus

#MultimodalUnderstanding #UnifiedFoundationModels #MultimodalDiffusionTransformers #SpatialIntelligence #MultimodalGeneration

🔥 OpenSearch-VL: An Open Recipe for Frontier Multimodal Search Agents

💡 The paper introduces OpenSearch-VL, an open-source framework for training advanced multimodal search agents using reinforcement learning. The problem addressed is that current top-tier multimodal search agents are difficult to reproduce due to the lack of open high-quality training data, transparent trajectory synthesis pipelines, and detailed training recipes. To solve this, the authors propose a fully open-source recipe for training frontier multimodal deep search agents.

The method involves curating high-quality training data through a dedicated pipeline that includes Wikipedia path sampling, fuzzy entity rewriting, and source-anchor visual grounding. This pipeline is used to create two training datasets, SearchVL-SFT-36k and SearchVL-RL-8k. The authors also design a diverse tool environment that combines text search, image search, and other tools to enable agents to acquire external knowledge.

A new training algorithm, multi-turn fatal-aware GRPO, is proposed to handle cascading tool failures by masking post-failure tokens while preserving useful pre-failure reasoning. The results show that OpenSearch-VL delivers substantial performance gains, with over 10-point average improvements across seven benchmarks, and achieves results comparable to proprietary commercial models on several tasks. The authors will release all data, code, and models to support open research on multimodal deep search agents.

Overall, the paper contributes to the development of multimodal search agents by providing an open-source framework, high-quality training data, and a novel training algorithm, making it easier for researchers to reproduce and improve upon the results.

📅 Published on May 6

🔗 Links:
• arXiv: https://arxiv.org/abs/2605.05185
• PDF: https://arxiv.org/pdf/2605.05185
• Project Page: https://huggingface.co/OpenSearch-VL
• GitHub: https://github.com/shawn0728/OpenSearch-VL ⭐ 81

🤖 Models citing this paper:
• https://huggingface.co/OpenSearch-VL/OpenSearch-VL-8B
• https://huggingface.co/OpenSearch-VL/OpenSearch-VL-30B-A3B
• https://huggingface.co/OpenSearch-VL/OpenSearch-VL-32B

📊 Datasets citing this paper:
• https://huggingface.co/datasets/OpenSearch-VL/Search-VL-RL-8K
• https://huggingface.co/datasets/OpenSearch-VL/Search-VL-SFT-36K

━━━━━━━━━━━━━━━━━━━━━━━━
📢 By: https://xn--r1a.website/PaperNexus

#MultimodalSearchAgents #ReinforcementLearningForSearch #OpenSourceSearchFrameworks #MultimodalDeepLearning #ReinforcementLearningForMultimodalSystems

🔥 Continuous-Time Distribution Matching for Few-Step Diffusion Distillation

💡 The paper introduces Continuous-Time Distribution Matching, a new method for accelerating diffusion models through distillation. The existing Distribution Matching Distillation method has limitations, as it relies on sparse supervision at discrete timesteps, leading to visual artifacts and over-smoothed outputs. To address this, the authors propose a continuous-time optimization approach that replaces the fixed discrete schedule with a dynamic continuous schedule, allowing distribution matching to be enforced at arbitrary points along sampling trajectories. Additionally, they introduce a continuous-time alignment objective that performs active off-trajectory matching on latents extrapolated via the student's velocity field, improving generalization and preserving fine visual details. The results show that this new method, called Continuous-Time Distribution Matching, provides highly competitive visual fidelity for few-step image generation without relying on complex auxiliary objectives, outperforming existing methods on various architectures. The code for the method is made available, demonstrating the effectiveness of this new approach for diffusion model distillation.

📅 Published on May 7

🔗 Links:
• arXiv: https://arxiv.org/abs/2605.06376
• PDF: https://arxiv.org/pdf/2605.06376
• Project Page: https://byliutao.github.io/cdm_page/
• GitHub: https://github.com/byliutao/cdm ⭐ 22

🤖 Models citing this paper:
• https://huggingface.co/byliutao/stable-diffusion-3-medium-turbo
• https://huggingface.co/byliutao/Longcat-Image-Turbo

━━━━━━━━━━━━━━━━━━━━━━━━
📢 By: https://xn--r1a.website/PaperNexus

#DiffusionModelDistillation #ContinuousTimeDistributionMatching #FewStepDiffusion #DistributionMatchingDistillation #DiffusionBasedImageSynthesis

🔥 MARBLE: Multi-Aspect Reward Balance for Diffusion RL

💡 The paper introduces MARBLE, a novel gradient-space optimization framework for multi-reward reinforcement learning fine-tuning of diffusion models. The problem addressed is that existing methods for handling multiple rewards either train separate models for each reward or use a weighted-sum reward aggregation, which can lead to poor performance due to sample-level mismatch. This mismatch occurs because most rollouts are highly informative for certain reward dimensions but irrelevant for others, causing the weighted summation to dilute their supervision.

To address this issue, MARBLE maintains independent advantage estimators for each reward and computes per-reward policy gradients. These gradients are then harmonized into a single update direction without manual reward weighting, by solving a quadratic programming problem. This approach allows for a unified model that can be jointly trained on all rewards, eliminating the need for heavy manual tuning and sequential training.

The authors also propose an amortized formulation that reduces the computational cost of MARBLE, making it more efficient. Additionally, they use exponential moving average smoothing on the balancing coefficients to stabilize updates against transient fluctuations.

The results show that MARBLE improves all five reward dimensions simultaneously on the SD3.5 Medium dataset, outperforming the baseline method. Specifically, MARBLE turns the worst-aligned reward's gradient cosine from negative to consistently positive, indicating better alignment with human preferences. Furthermore, MARBLE runs at nearly the same training speed as the baseline method, with only a 3% slowdown. Overall, MARBLE provides a more effective and efficient approach to multi-reward reinforcement learning fine-tuning of diffusion models.

📅 Published on May 7

🔗 Links:
• arXiv: https://arxiv.org/abs/2605.06507
• PDF: https://arxiv.org/pdf/2605.06507
• Project Page: https://aim-uofa.github.io/MARBLE/
• GitHub: https://github.com/aim-uofa/MARBLE ⭐ 24

━━━━━━━━━━━━━━━━━━━━━━━━
📢 By: https://xn--r1a.website/PaperNexus

#MultiRewardReinforcementLearning #DiffusionModels #GradientSpaceOptimization #MultiAspectRewardBalance #ReinforcementLearningFineTuning

🔥 SwiftI2V: Efficient High-Resolution Image-to-Video Generation via Conditional Segment-wise Generation

💡 The paper proposes SwiftI2V, an efficient framework for high resolution image to video generation. The problem addressed is that existing methods for generating high resolution videos from images are either computationally expensive or lack fidelity to the input image. High resolution image to video generation aims to synthesize realistic temporal dynamics while preserving fine grained appearance details of the input image, but at 2K resolution, this becomes extremely challenging. Existing solutions suffer from weaknesses such as high memory and latency costs, or hallucinating details and drifting from input specific local structures.

The proposed method, SwiftI2V, addresses these weaknesses by using a two stage design. First, it generates a low resolution motion reference to reduce token costs and ease the modeling burden. Then, it performs a strongly image conditioned 2K synthesis guided by the motion to recover input faithful details with controlled overhead. To make generation more scalable, SwiftI2V introduces Conditional Segment wise Generation, which synthesizes videos segment by segment with a bounded per step token budget. It also adopts bidirectional contextual interaction within each segment to improve cross segment coherence and input fidelity.

The results show that SwiftI2V achieves performance comparable to end to end baselines while reducing total GPU time by 202 times on the VBench I2V benchmark at 2K resolution. This enables practical 2K image to video generation on a single datacenter GPU or consumer GPU, making it a significant contribution to the field of image to video generation. Overall, SwiftI2V provides an efficient and scalable solution for high resolution image to video generation, addressing the efficiency fidelity dilemma and achieving state of the art results.

📅 Published on May 7

🔗 Links:
• arXiv: https://arxiv.org/abs/2605.06356
• PDF: https://arxiv.org/pdf/2605.06356
• Project Page: https://hkust-longgroup.github.io/SwiftI2V/
• GitHub: https://github.com/HKUST-LongGroup/SwiftI2V ⭐ 13

━━━━━━━━━━━━━━━━━━━━━━━━
📢 By: https://xn--r1a.website/PaperNexus

#ImageToVideoGeneration #HighResolutionVideoSynthesis #ConditionalSegmentWiseGeneration #EfficientVideoGeneration #ImageBasedVideoRendering