cập nhật 1 giờ trước

⏱ Độ trễ thật của tab này: trung bình 6.8 giờ sau khi CVE được công bố (n = 5262, sớm nhất 0.1 giờ). Số âm nghĩa là ta thấy trước ngày công bố.

20
Glance: The Generator Is Its Own Reranker
📄 Paper HF Daily Papers 22 ngày trước

Retrieval-augmented generation commonly serves separate models for reranking and answering, duplicating model residency and passage processing. We introduce Glance, which equips the generator with a lightweight decision adapter and pointer head to score candidate passages from a …

🅲 hạng C
20
Training Foundation Models with a Single Algorithm
📄 Paper HF Daily Papers 22 ngày trước

Foundation models are typically trained in stages: pretraining, midtraining, supervised fine-tuning (SFT), and reinforcement learning (RL), with changes in both data and learning algorithms. We show that this algorithmic separation is not fundamental. By deriving suitable RL obje…

🅲 hạng C
20
WorldBridge: Bridging Video and Action for Task Generalization in World Action Models
📄 Paper HF Daily Papers 22 ngày trước

Achieving the broad task generalization of embodied models in physical environments remains a central challenge in robot learning. We study this challenge in World Action Models (WAMs) through two complementary capabilities: video generalization, which predicts the visual evoluti…

🅲 hạng C
20
Human-to-Robot Transfer via Bridging Representation and Action Space in VLAs
📄 Paper HF Daily Papers 17 ngày trước

Pretrained Vision-Language-Action (VLA) policies provide a strong starting point for deployment, but adapting them to a target robot and environment still requires costly embodiment-specific demonstrations. Human demonstrations are far easier to collect, yet directly leveraging t…

🅲 hạng C
20
PyRUA-REAL: Robot Exploration and Adaptive Learning through RSI in the Real World
📄 Paper HF Daily Papers 15 ngày trước

Coding agents powered by models such as GPT-6-Astra connect spatial reasoning with executable robot behavior, raising the question of how such agents can produce lasting improvements in robot capabilities from physical interaction. In this work, we study embodied recursive self-i…

🅲 hạng C
20
Muir: Linear Attention with Sensitivity-Allocated Gating and Matched-Filter Readout
📄 Paper HF Daily Papers 14 ngày trước

Linear attention compresses context into a fixed-size key-value state for linear-time training and constant-memory inference. Gated delta-rule models decay this state and erase its prediction for an incoming key before writing a new value; Gated DeltaNet-2 (GDN-2) generalizes thi…

🅲 hạng C
20
ProTEUS: A mutation-resolved world model for closed-loop protein evolution
📄 Paper HF Daily Papers 11 ngày trước

Protein engineering requires navigating rugged fitness landscapes, where small sequence changes can produce large functional differences, but experimental screening can only sample a small fraction of sequence space. Thus, a central challenge is to efficiently traverse the large …

🅲 hạng C
20
UniSpark: Exploring Hybrid Attention for Multimodal Understanding and Generation
📄 Paper HF Daily Papers 8 ngày trước

We present UniSpark-9B, a unified multimodal model that investigates hybrid attention for visual understanding and generation. Built on a Mixture-of-Transformer-Experts backbone with approximately 9B parameters in each branch, UniSpark combines full attention for explicit token i…

🅲 hạng C
20
A frontier machine intelligence for human neuroimaging
📄 Paper arXiv cs.AI 6 ngày trước cũng có trên: HF Daily Papers

arXiv:2611.15094v1 Announce Type: new Abstract: Medical imaging is essential for diagnosing disease, guiding treatment and monitoring patients over time. Although artificial intelligence has advanced image interpretation, existing medical models often address restricted tasks, w…

🅲 hạng C
20
WebArena-Pro: A Heterogeneous, Multimodal, Reproducible Benchmark for Web Agents
📄 Paper HF Daily Papers 7 ngày trước

Web agents powered by large language and vision-language models are increasingly applied to realistic browser work that spans heterogeneous applications, multimodal content, and stateful workflows. However, existing reproducible web-agent benchmarks cover only a small number of w…

🅲 hạng C
20
DirCast: Live Esports Directing and Commentary with a Single Streaming Video-Language Model
📄 Paper HF Daily Papers 29 ngày trước

A live esports broadcast needs a director who chooses which camera goes on air and a caster who talks over that choice, every second and without knowing what comes next. Streaming video language models have learned to commentate, but only over a broadcast that a human director ha…

🅲 hạng C
20
RoboJev: An Open Vision-Language Judge for Robot Success Detection
📄 Paper HF Daily Papers 27 ngày trước

Many robot manipulation videos do not indicate whether the task was successfully completed. RoboJev-27B is an open-weight vision-language model that takes a task instruction, 10 frames of a manipulation attempt, and (if available) final frames from auxiliary cameras and predicts …

🅲 hạng C
20
SO-101 Bench: Accessible Spatial Manipulation
📄 Paper HF Daily Papers 24 ngày trước

Low-cost arms such as the SO-101 have become the most common entry point to robot learning, yet no released policy performs spatially constrained, language-conditioned tasks on them out of the box, and existing SO-101 benchmarks run only on physical hardware with simple tasks and…

🅲 hạng C
20
RynnWorld-Latent: A Cross-Embodiment World Model Conditioned on Universal Latent Actions
📄 Paper HF Daily Papers 24 ngày trước

Learning controllable world dynamics from heterogeneous interaction videos is challenging because robot embodiments use incompatible action spaces, while human videos lack native control labels. We present RynnWorld-Latent, a cross-embodiment video world model that uses the 3D-gr…

🅲 hạng C
20
The Lexical Imprint of Pre-training Data
📄 Paper HF Daily Papers 25 ngày trước

How much of zero-shot benchmark accuracy is captured by word-level alignment between a pre-training corpus and the benchmark target? We study this question through word-level unigram cross-entropy (UCE), a tokenizer-agnostic, model-free measure of lexical alignment between a refe…

🅲 hạng C
20
When Requirements Interact: Scaling Long-Horizon Coding via Interdependent Specifications
📄 Paper HF Daily Papers 22 ngày trước

To succeed in real-world long-horizon software development, coding agents must satisfy multi-dimensional task specifications, such as interface compatibility and error handling, and do so without breaking existing dependencies. However, current works lack specification in both qu…

🅲 hạng C
20
BYOD: Build Your Own DLM - Efficient Autoregressive-to-Diffusion Language Model Conversion
📄 Paper HF Daily Papers 21 ngày trước

Unlike autoregressive (AR) models, diffusion language models (DLMs) permit parallel token updates, flexible generation, and bidirectional attention, but leading systems rely on costly pretraining or full-model fine-tuning. We present Build Your Own DLM (BYOD), a generalizable LoR…

🅲 hạng C
20
Future Representation Distillation for Imagination-free World Action Models
📄 Paper HF Daily Papers 16 ngày trước

World Action Models (WAMs) use the predictive knowledge of pretrained video diffusion models for robot control through future video prediction. However, iterative video denoising is costly, and inaccurate visual predictions can compromise action prediction. Our key idea is to mak…

🅲 hạng C
20
HeLM: Hierarchical Explicit Language Memory for Non-Markovian Manipulation
📄 Paper HF Daily Papers 17 ngày trước

Recent advances in Vision-Language-Action (VLA) models have enabled fine manipulation capabilities, yet performing tasks that require working memory remains a key challenge. Most existing VLA architectures operate under a Markovian assumption, mapping only the current observation…

🅲 hạng C
20
Exploring Weight-Space Programming: Capabilities, Generalization, and Limitations
📄 Paper HF Daily Papers 14 ngày trước

Weight-Space Programming (WSP) enables reasoning agents to directly inspect and modify neural network parameters, offering an alternative to gradient-based model adaptation. However, its capabilities, cross-task generalization, and practical limitations remain insufficiently unde…

🅲 hạng C
20
How Many Exposures Does a World Model Need? Learning Unseen Game Mechanics at Test Time
📄 Paper HF Daily Papers 11 ngày trước

World models enable agents to plan and learn from imagined experience, but in an unfamiliar environment they must learn its mechanics from limited interaction. A mechanic is a rule for how an action or an interaction changes part of the environment. Each occurrence of a mechanic …

🅲 hạng C
20
How Much Training-Inference Mismatch Can On-Policy Distillation Tolerate?
📄 Paper HF Daily Papers 12 ngày trước

On-policy distillation (OPD) trains a student language model on its own responses, guided by a fixed teacher. Generating these responses is expensive, so systems use quantized inference and asynchronous rollouts to reduce the cost. Both, however, introduce training-inference mism…

🅲 hạng C
20
CANTraj: Recovering Vehicle Trajectories from CAN Signals in Naturalistic Driving
📄 Paper HF Daily Papers 7 ngày trước

Open driving releases and usage-based insurance programs often claim to prevent privacy breaches by omitting GPS records when sharing or collecting vehicle data. However, removing GPS coordinates from driving logs does not erase the location information carried by vehicle signals…

🅲 hạng C
20
Emergent Biological Capabilities in a Foundation Model for Molecular Interactions
📄 Paper HF Daily Papers 10 ngày trước

Foundation models trained on large unlabeled corpora develop emergent capabilities, tasks that go beyond the training objective. We extend this paradigm to multimodal biology with Bonbon, a foundation model for molecular interactions trained on approximately 3.2 trillion tokens o…

🅲 hạng C
20
HALO: Human-Aligned Data Collection for WhoLe-body MObile Manipulation
📄 Paper HF Daily Papers 8 ngày trước

Learning whole-body mobile manipulation requires demonstrations that coordinate bimanual actions with locomotion across extended workspaces. However, existing interfaces trade off direct robot control, perception and action alignment, and natural operator mobility, making such de…

🅲 hạng C
20
Can Test-Time Memory Evolution Outperform Test-Time Training for Scientific Discovery?
📄 Paper arXiv cs.AI 24 ngày trước cũng có trên: HF Daily Papers

arXiv:2610.32046v1 Announce Type: new Abstract: Recent \emph{test-time training} (TTT) frameworks have enabled open-weight large language models (LLMs) to outperform inference-only frameworks powered by frontier models on scientific discovery tasks. However, TTT introduces addit…

🅲 hạng C
20
Trust Mechanisms for Offline Generative AI at the Edge: A Connectivity-Regime Framework and an Empirical Study on a Raspberry Pi 5
📄 Paper arXiv cs.CR 23 ngày trước cũng có trên: HF Daily Papers

arXiv:2610.33732v1 Announce Type: new Abstract: Generative AI and large language models (LLMs) are moving into the Internet of Things, yet most deployments assume cloud connectivity, and many IoT settings have intermittent, restricted, or no connectivity. We propose a connectivi…

🅲 hạng C
20
Benchmarking Obstacle-Conditioned Safety in Household Robot Navigation
📄 Paper HF Daily Papers 25 ngày trước

Safe household navigation requires robots to regulate clearance and speed according to what an obstacle is and how close it is, without unnecessary conservatism. Yet existing evaluations do not test whether semantic safety distinctions are reflected in continuous motion. To that …

🅲 hạng C
20
Library of Layers (LoL): Adaptively Ordering Transformer Layers at Runtime
📄 Paper HF Daily Papers 17 ngày trước

Computer programmers chain together the same reusable computational primitives, for example, conditional statements or loops, but in different orders to implement different algorithms. In contrast, existing language model architectures always execute their layers in the same orde…

🅲 hạng C
20
Say the Same, Act Differently: Diagnosing and Steering Text-Action Divergence in Reasoning VLAs
📄 Paper arXiv cs.CR 15 ngày trước cũng có trên: HF Daily Papers

arXiv:2611.04166v1 Announce Type: cross Abstract: Reasoning vision-language-action (VLA) models pair physical control with natural-language rationales that supervisors can read to monitor embodied safety. Prior attacks on VLA reasoning corrupt the rationale itself and leave visi…

🅲 hạng C