arXiv Machine Learning
Jul 16
DarwinLM: Evolutionary Structured Pruning of Large Language Models
arXiv:2502. 07780v4 Announce Type: replace Abstract: Large Language Models (LLMs) have achieved significant success across various NLP tasks.
Large language models
Model releases, architecture work and prompting research on large language models — from frontier-lab announcements to the arXiv papers behind them.
arXiv Machine Learning
Jul 16
Inverting the Streaming-Diffusion Bottleneck: Video-Rate MLLM-Conditioned Edit Diffusion on a Consumer GPU
arXiv:2606. 05981v2 Announce Type: replace-cross Abstract: Aggressive distillation of the diffusion U-Net inverts the per-frame bottleneck of real-time text-to-image pipelines: once the denoiser is a 4-step or 1-step distilled student, the text encoder becomes the critical path.
arXiv Machine Learning
Jul 16
Diagnosing and Calibrating Tool-Call Boundary Drift in Multi-Teacher On-Policy Distillation
arXiv:2607. 07050v2 Announce Type: replace-cross Abstract: Agentic language models must learn when to call tools, when to consume tool responses, and when to answer directly.
arXiv AI
Jul 16
Semantic Anchoring for Robotic Action Representations
arXiv:2607. 13597v1 Announce Type: cross Abstract: Vision-Language-Action (VLA) models inherit rich semantic representations from pretrained Vision-Language Models, yet fine-tuning on limited robot demonstrations degrades this structure and undermines generalization.
arXiv AI
Jul 16
How Far Can Root Cause Analysis Go on Real-World Telemetry Data?
arXiv:2607. 13548v1 Announce Type: new Abstract: Identifying root causes in production microservice failures requires reasoning over large-scale, multimodal telemetry spanning metrics, logs, and traces, a problem that has proved resistant to both classical and LLM-based approaches.
arXiv AI
Jul 16
Transforming Rank: How Architecture Navigates the Spectral Pathologies of Depth
arXiv:2607. 14018v1 Announce Type: cross Abstract: We investigate how each component of the Transformer feedforward block architecture design determines how much rank survives across depth at initialization.
arXiv AI
Jul 16
FixItFlow: Automated Troubleshooting Guide Generation from Cloud Incidents
arXiv:2607. 13035v1 Announce Type: cross Abstract: Cloud services experience frequent incidents that require rapid diagnosis and resolution.
arXiv AI
Jul 16
Deep Interaction: An Efficient Human-AI Interaction Method for Large Reasoning Models
arXiv:2607. 14049v1 Announce Type: new Abstract: The emergence of Chain-of-Thought (CoT) reasoning has significantly enhanced the ability of large language models (LLMs) to tackle complex, multi-step tasks.
arXiv Machine Learning
Jul 16
Targeted Recovery of Weight-Space Mechanisms From Neural Networks
arXiv:2607. 13047v1 Announce Type: new Abstract: Parameter decomposition (PD) decomposes neural networks into interpretable computational components that faithfully reflect the original network's operations.
Hugging Face Trending Papers
Jul 15
Dysco: Dynamic Subspace Boosting to Mitigate LoRA Interference in Federated Learning
Federated fine-tuning of large pre-trained models increasingly relies on Low-Rank Adaptation (LoRA) to reduce communication and computation, but heterogeneous clients can make adapter aggregation unstable. We identify the data-parameter interference as a geometric source of this instability.
Hugging Face Trending Papers
Jul 15
SD-MAR: Multi-image Analytical Reasoning via Synthetic Data and Reinforcement Learning
Vision Language Models (VLMs) demonstrate strong perceptual abilities but remain limited in tasks requiring analytical reasoning across multiple visual states, such as multi-image comparison, change detection, and multi-step visual inference. These capabilities are critical for real-world multimodal applications where reasoning must be grounded in systematic differences between visual contexts.
Hugging Face Trending Papers
Jul 15
Transforming Rank: How Architecture Navigates the Spectral Pathologies of Depth
We investigate how each component of the Transformer feedforward block architecture design determines how much rank survives across depth at initialization. We reinterpret skip connections and normalization, long understood as controlling magnitude, as mechanisms for preserving gradient rank across depth, since the very matrix multiplications and nonlinear activations that make the network expressive also reduce the rank.
Towards Data Science
Jul 15
Don’t Let Claude Grade Its Own Homework
Cross-provider PR review with Codex in GitHub Actions, and why a second opinion from a different lab beats any self-review The post Don’t Let Claude Grade Its Own Homework appeared first on Towards Data Science .
Hugging Face Trending Papers
Jul 15
RxBrain: Embodied Cognition Foundation Model with Joint Language-Visual Reasoning and Imagination
Embodied cognition requires agents to connect high-level task reasoning with the physical states to be achieved. We introduce Hy-Embodied-RxBrain, an embodied cognition foundation model with joint language-visual reasoning and imagination.
Hugging Face Trending Papers
Jul 15
SIVA-RL: Sensitivity-Invariance Visual Alignment for Multimodal Reinforcement Learning
Reinforcement learning with verifiable rewards (RLVR) drives multimodal reasoning, but answer-level correctness does not guarantee that a vision-language model grounds its predictions in visual evidence. Existing visual-intervention methods contrast policy behavior on original and modified images, yet assign supervision by the type of intervention rather than its observed effect.
Hugging Face Trending Papers
Jul 15
Closed-Loop Knowledge Dynamics: An Operational Framework for Saturation and Escape
Feedback-driven loops support iterative improvement in large language models, reinforcement learning, and autonomous discovery, yet their gains often diminish under repeated internal feedback. We study why closed-loop knowledge systems saturate and what external information can move them beyond their current attractors.
Hugging Face Trending Papers
Jul 15
Partially Correlated Verifier Cascades in LLM Harnesses: Concave Log-Odds, Polynomial Reliability, and Blind-Spot Ceilings
Serial verification gates are a core reliability primitive in LLM harnesses: a candidate answer is returned only if $k$ verifier calls all accept it. Under conditionally independent gates, the recent Odds Law (arXiv:2606.
Hugging Face Trending Papers
Jul 15
MxGPS: Multiplex Graph Transformers for a Power Grid Foundation Model
Single-task fine-tuning of graph neural networks (GNNs) for power grid problems exhibits a systematic failure mode: models that achieve the lowest in-distribution error degrade the most under topology shift. We term this topology overfitting: the tendency of task-specific gradient signals to encode relational structure particular to the training topologies rather than the underlying physics, causing models to fail on unseen grids despite strong in-distribution performance.
Hugging Face Trending Papers
Jul 15
AgentCompass: A Unified Evaluation Infrastructure for Agent Capabilities
As Large Language Models (LLMs) evolve into autonomous agents, the need for unified evaluation infrastructure becomes critical. However, current evaluation pipelines remain highly fragmented and tightly coupled, hindering reproducibility and causing redundant engineering.
Hugging Face Trending Papers
Jul 15
Fine-grained CLIP fine-tuning with self-annotated region alignment
Contrastive Language-Image Pre-training (CLIP) has been shown to have limitations in its fine-grained dense feature representation, due to its pre-training focusing on matching the whole image to a text description. Considering the large data and computational burden in pre-training a vision-language model from scratch, a series of works aim to enhance the fine-grained ability of CLIP through a fine-tuning scheme.