cập nhật 1 giờ trước

⏱ Độ trễ thật của tab này: trung bình 6.9 giờ sau khi CVE được công bố (n = 5187, sớm nhất 0.1 giờ). Số âm nghĩa là ta thấy trước ngày công bố.

20
TorchCurves: Differentiable Parametric Curves
📄 Paper HF Daily Papers 10 ngày trước

We present TorchCurves, a PyTorch library that treats vector-valued parametric curves as learnable primitives. Its central operation maps batches of scalar inputs to points on curves represented in B-spline or Legendre polynomial bases, with differentiation through both inputs an…

🅲 hạng C
20
Visual Similarity in Adult Word Extension and Artificial Word Learning
📄 Paper HF Daily Papers 10 ngày trước

When two objects share a name, how much does their visual similarity explain, and what do the teaching examples add? First, we reanalyzed trial-level data from a large adult word-extension study (Lewis & Frank, 2018), computing each test image's distance to the examples shown in …

🅲 hạng C
20
Mixed Signals: Weight-Space Programming for EEG and ECG
📄 Paper HF Daily Papers 10 ngày trước

Weight-space programming allows language models to read a trained neural network's weights and edit them. This opens the door to ``training'' by reasoning rather than through gradient descent. Prior work has demonstrated its functionality and promise for game-playing networks and…

🅲 hạng C
20
Probing Transformer Representations Layer by Layer, Module by Module
📄 Paper HF Daily Papers 12 ngày trước

Which representation should be probed in a transformer-based foundation model? Recent work shows that the standard practice of probing the final layer is often suboptimal, as intermediate layers can provide better features. Yet, representations are still largely studied along dep…

🅲 hạng C
20
The Narrative Gap: Can LLMs Help Us Navigate Diverse Narratives Across Languages?
📄 Paper HF Daily Papers 11 ngày trước

Large language models (LLMs) increasingly mediate how people seek and consume information, including accounts that differ across languages and regions, especially on contested topics. Supporting diverse information seeking requires models to help users navigate competing narrativ…

🅲 hạng C
20
Iris-Z: Scaling One Generative Model Past Specialist Geometry Models
📄 Paper HF Daily Papers 9 ngày trước

Language moved from many task-specific models to one unified model: a single generative model with a single generative training task that handles all of them within the same backbone, and keeps improving with data, parameters and compute. We believe the bitter lesson applies just…

🅲 hạng C
20
Dataset Distillation via Drifting
📄 Paper HF Daily Papers 8 ngày trước

The task of dataset distillation concerns the learning of a small set of synthetic samples that can serve as an effective replacement for a large real dataset when used for downstream tasks. While many solutions have been proposed since the problem's inception, most of them prima…

🅲 hạng C
20
Skeleton-Grounded Latent Action Model for Robot World Models
📄 Paper HF Daily Papers 29 ngày trước

Robot world models simulate how environments evolve in response to robot actions. Latent action models enable learning dynamics from videos by encoding visual transitions as proxy actions. However, reconstruction-driven latent actions mix robot motion with object responses and ot…

🅲 hạng C
20
HoloBoxing: A Simple Recipe for Humanoid Boxing with 7 Minutes of Single-GPU Training
📄 Paper HF Daily Papers 28 ngày trước

HoloBoxing presents a simple recipe for learning humanoid boxing by building on a pretrained whole-body controller. The recipe separates reusable motor skills from task decisions: a compact planner requests future body-point targets, while the frozen controller handles joint-leve…

🅲 hạng C
20
A path-following framework on fiber bundle for variational inequalities
📄 Paper HF Daily Papers 28 ngày trước

Developing general-purpose algorithms with global asymptotic convergence guarantees for variational inequalities (VIs) remains a longstanding challenge. This paper proposes a path-following framework for finite-dimensional VIs with arbitrary continuous mappings and polytope domai…

🅲 hạng C
20
The fixed-point bundle method over product-of-simplex domains arising from game equilibria
📄 Paper HF Daily Papers 28 ngày trước

Developing general-purpose algorithms with global asymptotic convergence guarantees for Nash equilibria of general-sum games remains a longstanding challenge. This paper extends the fixed-point bundle framework for finite-dimensional variational inequalities (VIs) from simplex do…

🅲 hạng C
20
Does LeWorldModel Preserve Local Action Geometry?
📄 Paper HF Daily Papers 29 ngày trước

We examine how closely pretrained LeWorldModel (LeWM) reproduces the effects of small action changes. Using matched simulator and model responses, we measure action sensitivity with Gram matrices at three perturbation scales. Across nearly a thousand states from Reacher and OGBen…

🅲 hạng C
20
CellTrack-Bench: Microscopic Cell Tracking Benchmark for Vision-Language Models
📄 Paper HF Daily Papers 29 ngày trước

Large language models have shown outstanding performance on text-first tasks such as coding, math, and writing. Recent efforts are shifting models' training to include scientific problems that often live in highly specialized domains beyond text, such as protein structure and bio…

🅲 hạng C
20
Harness-Model Collaboration for Long-Horizon Continual Learning
📄 Paper HF Daily Papers 25 ngày trước

Long-horizon agents continuously encounter reusable procedural knowledge, yet existing systems typically rely either on model adaptation or external memory, trading persistent learning for rapid reuse. This limitation is especially important for Web agents that repeatedly execute…

🅲 hạng C
20
PhysEvolve: Verifier-grounded Self-evolution for Physical Reasoning with Executable Worlds
📄 Paper HF Daily Papers 25 ngày trước

Physical reasoning underpins how embodied systems understand and act in the physical world, yet vision-language models (VLMs) remain unreliable on questions about dynamics and physical causality. Progress is critically constrained by the scarcity of grounded counterfactual traini…

🅲 hạng C
20
HuGe: Scaling Humanoid Generalist Vision-Language-Action Models with Egocentric Experience
📄 Paper HF Daily Papers 26 ngày trước

Extending Vision-Language-Action models (VLAs) to humanoid robots requires coordinated whole-body control and diverse demonstrations that are costly to collect in real-world scenarios. Although egocentric human demonstrations provide a complementary source of task experience, dif…

🅲 hạng C
20
Instruction Capacity: Behavior and Degradation of Language Models under Controlled Load
📄 Paper HF Daily Papers 25 ngày trước

The quality with which language models follow instructions falls as the load grows, yet there is still no quantity that expresses the limit of a model in units of load with an uncertainty. We propose an instrument in which the load consists of rules that replace single-token word…

🅲 hạng C
20
Know, then Draw: A New Learning Paradigm for Visual Generation
📄 Paper HF Daily Papers 24 ngày trước

Visual generation requires both high-level semantic knowledge and fine-grained pixel details. We propose a learning paradigm that assigns these roles to autoregressive pretraining for knowledge acquisition and diffusion post-training for high-fidelity rendering. Within this frame…

🅲 hạng C
20
PreView: Enhancing GUI Grounding with Preprocessed Views and Cues
📄 Paper HF Daily Papers 27 ngày trước

GUI grounding requires locating small controls in cluttered screenshots and identifying which control a natural-language instruction refers to. Cropping and enlargement can expose local detail, but the resulting image may still leave ambiguity about the intended target. We propos…

🅲 hạng C
20
FDSML: Formula-Driven Supervised Metric Learning with Parameter-Aware Triplet Loss
📄 Paper HF Daily Papers 22 ngày trước

Formula-Driven Supervised Learning (FDSL) enables vision pre-training without real images, but existing methods cast pre-training as classification on discretized labels, discarding continuity in the generative process. We propose Formula-Driven Supervised Metric Learning (FDSML)…

🅲 hạng C
20
TurnTurk-GPT: Text-Based Turn-Taking Prediction for Turkish
📄 Paper HF Daily Papers 21 ngày trước

This paper presents TurnTurk-GPT, a text-based turn-taking predictor for Turkish obtained by fine-tuning a Turkish GPT-2 model with the TurnGPT objective on synthetic and real dialogue, and evaluates it on real Turkish conversation against rule baselines, a zero-shot Turkish lang…

🅲 hạng C
20
Think Twice: The Metis-1.6 Technical Report
📄 Paper HF Daily Papers 24 ngày trước

Metis is a bet that better data, better training methods and new architectures can buy what a bigger parameter count cannot, at a fraction of the cost. Metis-1.6 is its latest step: two language models built from scratch by a single researcher, Praxis (2.94B parameters, 0.46B act…

🅲 hạng C
20
On-Policy Self-Distillation with Teacher Evolution
📄 Paper HF Daily Papers 23 ngày trước

On-Policy Self-Distillation (OPSD), which aligns a student model's token-level distribution with that of its self-teacher prompted by privileged context, has emerged as a promising direction to mitigate reward sparsity in sequence-level post-training paradigms. However, OPSD ofte…

🅲 hạng C
20
TaoMate-FVSR:Audio-Guided Facial Video Restoration
📄 Paper HF Daily Papers 18 ngày trước

AI-generated videos often depict people within detailed scenes while leaving their small facial regions poorly resolved. Visual super-resolution can sharpen these regions without resolving whether the recovered lips and teeth agree with the spoken audio. We present an audio-guide…

🅲 hạng C
20
Where to Learn and How Far to Roll Out: Efficient Adaptive Training for GUI Agents
📄 Paper HF Daily Papers 22 ngày trước

Trajectory-level GRPO is a common approach for training multi-step GUI agents, but repeatedly rolling out complete trajectories from the task root can be inefficient: task-level rewards provide coarse supervision over which intermediate decisions require improvement, while alread…

🅲 hạng C
20
SeMo3D: Unifying 3D Generation, Editing, and Completion via Selective Modeling
📄 Paper HF Daily Papers 18 ngày trước

3D generation, editing, and completion are typically treated as regenerating the complete 3D asset under task-specific constraints, even when only a local region needs modification. This leads to task isolation, inconsistency in content that should remain unchanged, and dependenc…

🅲 hạng C
20
Is Forgetting Unified? Exploring Concept Unlearning in Unified Multimodal Models
📄 Paper HF Daily Papers 17 ngày trước

Unified multimodal models (UMMs) integrate image generation and visual understanding, but are existing single-pathway unlearning methods sufficient to suppress target concepts across both? In this work, we study cross-pathway concept unlearning. Our goal is to prevent a model fro…

🅲 hạng C
20
Every Pixel Is an Option: A System One Model for Medical Ultrasound Biometry
📄 Paper HF Daily Papers 18 ngày trước

**Purpose:** Ultrasound biometry is widely used across various clinical applications, including prenatal and cardiac examinations, to obtain quantitative measurements by placing calipers at anatomical landmarks. However, these measurements remain subject to interoperator variabil…

🅲 hạng C
20
BEND 2.0: Benchmarking the Capabilities Needed to Design DNA
📄 Paper HF Daily Papers 17 ngày trước

Genomic foundation models are increasingly used to propose DNA sequences and edits. But successful design requires capabilities that current benchmarks do not measure. Here we introduce BEND 2.0 (BENchmark for DNA language models), which tests how well models predict the effects …

🅲 hạng C
20
Sketchpad: certified exact arithmetic with a small vision-language model
📄 Paper HF Daily Papers 17 ngày trước

Language models are accurate on most long-arithmetic problems, but their accuracy is estimated on samples and carries no guarantee for an unseen input; tool use is exact but moves the computation outside the network. We show that exactness at all input lengths becomes a finite, c…

🅲 hạng C
20
InstantHMR : Real-Time Human Mesh Recovery on Your Phone
📄 Paper HF Daily Papers 14 ngày trước

Real-time 3D human mesh recovery from a single image on mobile devices is a challenging problem with many applications, including robotics, biomechanics, AR/VR, and exergames. Current state-of-the-art models are either too slow to run efficiently on mobile devices or sacrifice ac…

🅲 hạng C
20
Creative Priming: Decoupling Process from Product to Escape the Creative Paradox
📄 Paper HF Daily Papers 11 ngày trước

Many of the most consequential advances promised by AI, from curing diseases and inventing new technologies to formulating new scientific theories, are fundamentally creative acts. Yet creativity is uniquely difficult to learn because it faces a creative paradox: learning to repe…

🅲 hạng C
20
FraudBench: A Legal Evaluation of AI Deception
📄 Paper HF Daily Papers 11 ngày trước

We introduce FraudBench, an evaluation of large language models' (LLMs) propensities to recommend, assist, or directly take actions that would constitute fraud under U.S. law. Our evaluation contains a novel dataset of 159 scenarios based on real-world U.S. court cases and valida…

🅲 hạng C
20
RidgeFill: Training-Free Late-Layer Cache Fill for Faster Prefill
📄 Paper HF Daily Papers 14 ngày trước

RidgeFill shortens prefill by not computing most of a model's upper layers for the prompt. The lower layers run exactly up to a split; from the split layer's state a per-layer map predicts every later layer's input, and each late layer writes its cache from that prediction. The d…

🅲 hạng C