Hugging Face Daily Papers
500 articles archived · Visit source ↗ · RSS
-
-
Hugging Face Daily Papers research 5d ago
Hydra-0: Action Flow for Generalist World Modeling and Control
Abstract Hydra-0 uses action flow as a shared visual interface for generalist world modeling and robot control across diverse embodiments and tasks. Generated by thinkingmachines/Inkling-Small We introduce Hydra-0, a generalist world model conditioned on action flow, which…
21 -
Partition the Support, Reconstruct the Residual: Training-Free Sparse Attention for Video Generation and World Models
Abstract SparsePR accelerates video transformers via response-coupled partitioning and probe-fitted residual reconstruction, reducing attention error at low executed-pair densities with substantial speedups. Generated by thinkingmachines/Inkling-Small Training-free block-sparse…
33 -
ParaTempo: Efficient Parallel Reasoning via Temporal Confidence
Abstract ParaTempo improves parallel reasoning efficiency by using temporal confidence to dynamically prune, retire, and reallocate reasoning branches without synchronization. Generated by thinkingmachines/Inkling-Small Parallel reasoning improves the accuracy and robustness of…
18 -
FlavourBench: Ranking Frontier Language Models with Executable Culinary Ground Truth
Abstract FlavourBench evaluates language models on culinary portfolio tasks using executable ground truth, statistical rigor, and fully reproducible verification. Generated by thinkingmachines/Inkling-Small Open-ended language-model benchmarks usually inherit a judge: a human…
9 -
Human-Centric Intelligence in the Era of Foundation Models: A Survey
Abstract This survey proposes a unified taxonomy and methodological framework for human-centric intelligence across visual, dynamic, and embodied levels within the foundation-model era. Generated by thinkingmachines/Inkling-Small Human-centric intelligence is evolving in the…
19 -
UniSpace: Unified Visual Representation and Scalable Multimodal Modeling
Abstract A reparameterized pretrained vision transformer unifies semantic understanding, high-fidelity reconstruction, and image generation within a single visual space without requiring a separate VAE. Generated by thinkingmachines/Inkling-Small Semantic vision encoders have…
26 -
Daedalus-150M: A Convolution-Attention Hybrid Designed for CPU Inference
Abstract A small hybrid language model using sparse attention and short convolutions achieves faster CPU inference and better benchmark scores than larger conventional models despite training on far less data. Generated by thinkingmachines/Inkling-Small Small language models are…
26 -
EviRank: Structured Relevance Evidence for Multimodal Image Re-ranking
Abstract EviRank reformulates multimodal image re-ranking as semantic constraint satisfaction by parsing queries into structured evidence packages and verifying candidates via rubric scoring and listwise comparison without training. Generated by thinkingmachines/Inkling-Small…
35 -
Every Coin Has Two Sides: On the Dual Nature of Generalization in On-Policy Distillation of Large Language Models
Abstract On-policy distillation transfers reasoning behaviors rather than specific answers, with generalization strongly tied to teacher-student origin alignment and multi-teacher combinations causing capability trade-offs. Generated by thinkingmachines/Inkling-Small On-policy…
34 -
Llama-Mobile: Efficient 2.7-Bit Quantization of VLMs
Abstract A framework quantizes vision-language models for mobile deployment using self-generated training data and a 2.7-bit format, compressing Llama 3.2 11B Vision Instruct to 3.7 GB with preserved visual question answering performance. Generated by…
17 -
Let's Scale Step by Step: Compute-Efficient Hyperparameter Transfer for Large-Scale Mixture-of-Experts
Abstract A two-step hyperparameter transfer framework predicts optimal learning rates for large Mixture-of-Experts models by scaling across widths and token budgets, enabling efficient pretraining without costly sweeps. Generated by thinkingmachines/Inkling-Small…
19 -
OmniAssistBench: Assistant-style Interaction Benchmark for Omni-LLMs
Abstract OmniAssistBench evaluates real-time interactive video assistants by reverse-engineering multi-turn interaction videos, revealing that current omni-modal models struggle with visual prompts, context retention, and timely responses. Generated by…
33 -
AgentMercury: Your Agent Can Synthesize Verifiable Environments for Business Scenarios at scale
Abstract AgentMercury synthesizes scalable executable business environments that serve as generalizable reinforcement learning substrates, improving agent performance across enterprise and out-of-domain reasoning tasks while making environment construction itself learnable.…
38 -
Hadith computational science in the age of large language models: a critical narrative review
Abstract Hadith computational science is evaluated as an evidence infrastructure challenge requiring integration of transformer and retrieval-based methods with expert validation and provenance. Generated by thinkingmachines/Inkling-Small We examine how hadith computational…
38 -
Towards Faithful Simulation of Human Shopping Behavior
Abstract RecVerse is a GUI-grounded agent that uses hierarchical memory and trajectory-level reinforcement learning to simulate realistic multi-turn e-commerce shopping sessions. Generated by thinkingmachines/Inkling-Small Simulating realistic user shopping behavior underpins…
23 -
Graph Engineering in the Era of LLM Agents: From Individual Intelligence to System Intelligence
Abstract Graph Engineering organizes multi-agent LLM systems through dynamic graph structures to coordinate specialized agents and manage complex, evolving tasks. Generated by thinkingmachines/Inkling-Small LLMs have evolved from language generators to autonomous agents capable…
21 -
InfinityEdit: Infinite Video Editing with a Lightweight Edit-Ignition Adapter
Abstract InfinityEdit enables continuous, unbounded video editing by using a lightweight adapter with history, temporal causal, and edit cross-attention modules to extend edits into future streaming frames while maintaining stability. Generated by thinkingmachines/Inkling-Small…
20 -
CLEAR: Continuous Latent Adapter Routing for Utility-Preserving LLM Safety Alignment
Abstract CLEAR uses a hidden-state gate to continuously modulate a safety low-rank adapter, improving LLM safety while preserving utility on benign inputs. Generated by thinkingmachines/Inkling-Small Improving the safety of large language models (LLMs) often comes at the expense…
37 -
Beyond Correctness: Benchmarking and Aligning Response Behaviors in Hybrid-Thinking MLLMs
Abstract Hybrid-thinking multimodal language models suffer from response-pattern misalignment between thinking and non-thinking modes, which is addressed by a diagnostic benchmark and pattern-specific reinforcement learning penalties. Generated by thinkingmachines/Inkling-Small…
4 -
FlowEvo: Self-Evolving Agents through the Co-Evolution of Workflows and Executable Skills
Abstract FlowEvo enables large language model agents to co-evolve reusable skills and workflows during inference, improving accuracy and efficiency across diverse benchmarks. Generated by thinkingmachines/Inkling-Small Large language model agents can adapt to complex tasks by…
15 -
TinyCast: Probabilistic Zero-Shot Forecasting with Computed Periodicity
Abstract TinyCast is a compact, attention-free zero-shot forecaster that uses spectral period detection and dilated convolutions to emit predictive distributions with minimal parameters and embedded-device compatibility. Generated by thinkingmachines/Inkling-Small We introduce…
6 -
The Embedder's Dilemma: LLMs Are Better, but at What Cost?
Abstract Large language models and dedicated embedding models achieve nearly identical aggregate performance across diverse tasks, but embedding models are far cheaper and faster, supporting a division of labor by task type. Generated by thinkingmachines/Inkling-Small Should you…
33 -
τ_0-VLA: a Hierarchical Robot Foundation Model with World-Model-Guided Test-Time Computation
Abstract A hierarchical vision-language-action model improves long-horizon robot manipulation by using world-model-guided test-time search to scale computation for high-level subtask decisions. Generated by thinkingmachines/Inkling-Small Long-horizon robot manipulation requires…
5 -
QuoteBench: How Matched Scores Can Hide Command-Path Failures
Abstract QuoteBench reveals that execution-boundary parsing errors significantly reduce LLM coding agent success, and disclosing the boundary helps recover performance, showing that evaluation must account for deployment configuration rather than treating matched scores as…
18 -
Hierarchical Self-Improvement: A Framework for Task-Specific Evolvable Agent Harnesses
Abstract Hierarchical Self-Improvement evolves task-specific execution harnesses for frozen LLM agents via hierarchical self-modification, yielding substantial gains on moderate tasks while being bounded by feedback quality and backbone limits. Generated by…
30 -
GOAG: Generative and Object-Agnostic Grasp Planner for Dexterous Robotic Manipulation
Abstract GOAG is an object-agnostic deep generative grasp planner that learns a gripper-specific contact surface distribution to sample valid grasps for unseen objects without object-specific training. Generated by thinkingmachines/Inkling-Small Multifingered grasping is a…
35 -
CoToGrasp: Contact-Topology-Conditioned Dexterous Grasp Synthesis via Canonical Workspace Learning
Abstract CoToGrasp is a generative framework that synthesizes diverse, stable grasps conditioned on specific contact topologies using an object-agnostic, gripper-centric workspace for zero-shot generalization. Generated by thinkingmachines/Inkling-Small Current dexterous grasp…
22 -
Thinking in a Low-Resource Language: What SFT Builds, What RL Fixes, What Accuracy Cannot See
Abstract Fine-tuning large mixture-of-experts models on a low-resource language shifts reasoning into that language without harming accuracy, while reinforcement learning with verifiable rewards fixes formatting and leakage defects. Generated by thinkingmachines/Inkling-Small…
11 -
Listening Forward: Next Patch Embedding Prediction Enables Scalable Audio Learners
Abstract NAPE uses causal Transformers to predict successive spectrogram patch embeddings for self-supervised audio learning without auxiliary components. Generated by thinkingmachines/Inkling-Small Self-supervised learning (SSL) has driven substantial progress in audio…
6 -
NARU: A Benchmark for NARrative Evolution and Cultural Nuance Understanding in Japanese Extreme Long Video
Abstract NARU is a Japanese long-form video benchmark evaluating narrative evolution and cultural reasoning through a hierarchical annotation pipeline and extensive native-speaker verification. Generated by thinkingmachines/Inkling-Small Long-form video understanding encompasses…
35 -
EXIMO: VLM Guided Exploration of VLA Policies
Abstract EXIMO efficiently fine-tunes large vision-language-action robot policies by combining VLM-guided exploration, imitation on orchestrated data, and residual off-policy reinforcement learning. Generated by thinkingmachines/Inkling-Small How to efficiently finetune robot…
6 -
Towards Quantifying Benchmark Optimization in ASR Models
Abstract High-performing speech recognition models reproduce benchmark transcripts despite contradictory audio, revealing benchmark-optimized behaviors that inflate scores without improving real-world transcription. Generated by thinkingmachines/Inkling-Small Public benchmarks…
13 -
SWE-bench Science: Can Coding Agents Resolve Engineering Tasks in Science?
Abstract SWE-bench Science benchmarks coding agents on scientific software repair, revealing failure mechanisms and mixed effects of scientific guidance. Generated by thinkingmachines/Inkling-Small Software increasingly functions as part of the scientific instrument itself,…
17 -
Chain-of-Experience for Continual LLM Improvement
Abstract Large language models improve continuously through iterative test-time feedback loops called Chain-of-Experience, outperforming zero-shot baselines with lower cost and higher token efficiency. Generated by thinkingmachines/Inkling-Small Humans continuously learn from…
26 -
FlashPrefill V2: Block-Sparse Prefill Attention for Long-Context LLM Serving
Abstract FlashPrefill V2 improves long-context serving via mean-corrected sparse attention, optimized GPU operators, and framework integration, achieving large speedups over dense baselines. Generated by thinkingmachines/Inkling-Small Long-context modeling is a pivotal…
9 -
Repo0: Design-Driven Zero-to-All Code Generation
Abstract Repo0 uses a dual-graph architectural state and modularity-guided structural evolution to generate complete software repositories from natural-language requirements with high functionality coverage. Generated by thinkingmachines/Inkling-Small Large language model agents…
15 -
Inject, Align, Recover: Staged Post-Training for Retrieval-Free Document Knowledge Internalization
Abstract IAR is a three-stage post-training framework that injects structured document knowledge into language models, aligns them for retrieval-free question answering, and recovers general capabilities, improving both domain accuracy and general performance. Generated by…
7 -
EnvHarness: Awakening Static Worlds for Agent Learning
Abstract EnvHarness and EnvRigger dynamically reshape static environments via programmable plugins to target agent weaknesses and improve reinforcement learning co-evolution. Generated by thinkingmachines/Inkling-Small LLM agents learn by interacting with environments, yet these…
25 -
4DAnyone: Create Anyone in 4D from a Casual Monocular Video
Abstract 4DAnyone reconstructs 4D humans from monocular video by generating multiview-consistent videos and lifting them into 4D Gaussian Splatting, using reference and target context designs to overcome scaling bottlenecks. Generated by thinkingmachines/Inkling-Small We present…
24 -
MemTrapBench: Benchmarking Cognitive Traps in LLM Memory Use
Abstract Retrieved memories can induce reasoning errors and belief distortions in large language models, and an inference-time strategy helps avoid these cognitive traps while maintaining benchmark performance. Generated by thinkingmachines/Inkling-Small Memory has become a key…
12 -
ForgeWM: Progressive Causal Training for Few-Step Action-Conditioned Video World Models
Abstract ForgeWM progressively distills bidirectional video generators into efficient few-step interactive world models with aligned discrete and continuous controls, supporting low-latency interaction and replay-time refinement. Generated by thinkingmachines/Inkling-Small…
34 -
WithEveryone: Unified Planning and Identity Grounding for Group Image Generation
Abstract WithEveryone enables reliable identity-preserving group image generation for up to ten people by grounding identities to layout plans and using region-based identity losses. Generated by thinkingmachines/Inkling-Small Identity-preserving image generation becomes…
17 -
PolicyGuide: From Guarding One Action to Guiding the Whole Workflow for Policy-Compliant LLM Agents
Abstract Customer-service LLM agents must follow organizational policy when acting on a user's behalf. Compliance failures arise from either forbidden actions, such as granting an ineligible change, or omitted procedural requirements, such as identification or confirmation.…
22 -
VA-Judger: Reward Modeling from Human Preference Feedback for Joint Video-Audio Generation
Abstract A human-aligned chain-of-thought reward model and preference dataset improve joint video-audio generation by replacing fragmented metrics with coherent, dimension-wise reinforcement learning. Generated by thinkingmachines/Inkling-Small Using reinforcement learning to…
29 -
Evaluating Music Context Preservation: A Multi-facet Framework for Music Editing Systems
Abstract The study introduces MuseCPEval, a framework with tailored metrics to evaluate preservation of unchanged musical attributes during editing tasks. Generated by thinkingmachines/Inkling-Small Music editing plays a vital role in modern music production, with applications…
22 -
Towards Real-Time and Adaptable LiDAR Scene Completion
Abstract RapidLiDAR learns adaptive spatial displacements to initialize LiDAR scenes and refines them via multi-scale voxel and BEV features for real-time completion. Generated by thinkingmachines/Inkling-Small LiDAR scene completion is a key component of 3D perception in…
30 -
SPK: Eliciting Structured Prior Knowledge for Interpretable Out-of-Distribution Detection in Real-Time Object Detection
Abstract Structured Prior Knowledge explicitly extracts latent semantic, geometric, and contextual priors from pretrained object detectors to detect out-of-distribution hallucinations via a compact interpretable representation. Generated by thinkingmachines/Inkling-Small Object…
36 -
LLMs Get Smarter from Targeted Synthetic Multilingual Data
Abstract HOTFIXR is a data generation framework that targets multilingual reasoning weaknesses to improve cross-lingual performance without sacrificing overall capability. Generated by thinkingmachines/Inkling-Small Language-specific competency (LSC) is the phenomenon of a…
28 -
Bounded Agents: Delegation Security for Multi-Agent AI Systems
Abstract The Agentic Principal Chain enforces session-aware authorization checks to prevent harmful action combinations and delegation abuses in LLM agents. Generated by thinkingmachines/Inkling-Small LLM-based agents can act on behalf of a user to access cloud services, call…
36