cập nhật 1 giờ trước

⏱ Độ trễ thật của tab này: trung bình 6.9 giờ sau khi CVE được công bố (n = 5188, sớm nhất 0.1 giờ). Số âm nghĩa là ta thấy trước ngày công bố.

20
What Outcome RL Buys: Finishing, Reliability, Rarely Reach
📄 Paper HF Daily Papers 25 ngày trước

Outcome RL raises pass@1 but not pass@k at large k (Yue et al., 2025). But every pass@1 is scored under a token budget. Every recipe also trains under one. On physics and math problems at fixed token budgets, we compare the pass@1 of 20 released RL models (1.5B to 32B) with that …

🅲 hạng C
20
SciGame: Let Agents Play the Scientific Discovery Games in the Real and Counterfactual World
📄 Paper HF Daily Papers 23 ngày trước

Large language model agents are increasingly used for scientific discovery. However, existing benchmarks mix law inference with programming, score data fitting instead of prediction on unseen inputs, and do not separate empirical evidence from pretrained priors. We introduce SciG…

🅲 hạng C
20
Lab2Sim: Turning Real-World Experiment Videos into Interactive Virtual Experiments
📄 Paper HF Daily Papers 17 ngày trước

Science experiment videos demonstrate how experiments are performed, but offer no direct interaction for learners. Turning them into virtual experiments requires not only reproducing the observed apparatus and state changes, but also enforcing the preconditions governing each ope…

🅲 hạng C
20
Where Do Vision Language Models Fail at 3D Reasoning?
📄 Paper HF Daily Papers 22 ngày trước

Vision-language models can get individual judgments right while struggling to recover and consistently use the exact properties of a 3D object. To characterize this gap, we introduce BedBench, a controlled benchmark of thirteen procedurally generated tasks spanning individual jud…

🅲 hạng C
20
LiveFoley: Dynamic Real-Time Video-to-Audio Generation for Immersive Interactive Worlds
📄 Paper HF Daily Papers 16 ngày trước

Streaming video-to-audio (V2A) generation is essential for immersive interactive applications. However, real-time constraints leave existing streaming V2A methods far behind offline models: (1) visual frontends shrink their receptive fields to cut look-ahead latency, severely deg…

🅲 hạng C
20
Multi-layer Indexer: Overlapping Selection and Attention in DeepSeek Sparse Attention
📄 Paper HF Daily Papers 18 ngày trước

Sparse attention cuts the cost of long-context inference by making the attention matrix sparse: each query attends only a subset of the sequence, chosen by a fixed pattern or, in recent production LLMs, learned per query. In DeepSeek Sparse Attention (DSA), the attention of DeepS…

🅲 hạng C
20
StepX-OPD: Rethinking Multi-Teacher On-Policy Distillation in Diffusion Models
📄 Paper HF Daily Papers 17 ngày trước

A deployable text-to-image diffusion model must hold several capabilities at once, such as instruction following, text rendering and human preference, and must sample in a few steps. Capabilities and few-step sampling are built in separate stages: a full-step model first consolid…

🅲 hạng C
20
OmniLatent: Unified Latent for 3D Part Modeling
📄 Paper HF Daily Papers 14 ngày trước

3D part decomposition is a central capability for editable asset creation, yet existing workflows typically segment a completed mesh as a post-processing step, making the result sensitive to geometric artifacts, texture ambiguity, and limited part-level annotations. We present Om…

🅲 hạng C
20
Training Vision Transformers to Focus: Emergence of Spatial Attention Head Specialization
📄 Paper HF Daily Papers 11 ngày trước

Self-supervised learning in Vision Transformer models, such as DINO, enables robust visual perception without requiring labeled data by leveraging large-scale image datasets. However, these models lack a focusing mechanism during training, resulting in perceptual representations …

🅲 hạng C
20
ExRL: Sample-Efficient Online Adaptation of Long-Chunk Policies via Execution Length Control
📄 Paper HF Daily Papers 10 ngày trước

Predicting long action chunks, often spanning dozens of timesteps, has become a key ingredient in increasingly capable robot policies. In this paper, we find that adapting these long-chunk policies with existing reinforcement learning (RL) methods can require many online samples …

🅲 hạng C
20
Are Dense Rewards Worth Their Tokens in GRPO?
📄 Paper HF Daily Papers 10 ngày trước

RLVR trains language models to reason by rewarding correct final answers. A widely used algorithm, group-relative policy optimization (GRPO), gives every token of a solution the same credit, so dense process rewards have been proposed to credit intermediate steps, for example wit…

🅲 hạng C
20
Blind Variation, Selective Retention: Improving Creative Idea Generation in AI Scientists
📄 Paper HF Daily Papers 11 ngày trước

Scientific discovery requires ideas that are original, useful, and non-obvious. Yet AI scientists are typically given as much task information as possible during idea generation, which can anchor generation toward familiar approaches. Inspired by the cognitive framework of Blind …

🅲 hạng C
20
From Large Language Models to Efficient System One Models via Adaptive Computation
📄 Paper HF Daily Papers 8 ngày trước

Transferring LLMs into System One decision models typically replaces the language modeling head with a decision head while retaining the full backbone, making every decision pay for all layers. Yet it remains unclear how much depth such decisions require. We find that decision in…

🅲 hạng C
20
EgoX: Egocentric Cross-Embodiment Manipulation with Embodiment Dreaming
📄 Paper HF Daily Papers 8 ngày trước

Generalist manipulation policies can benefit from the complementary experience of humans, real robots, and simulated agents. However, differences in morphology, sensing, and control interfaces make it difficult to learn effectively from their combined demonstrations. We introduce…

🅲 hạng C
20
A frontier machine intelligence for human neuroimaging
📄 Paper HF Daily Papers 9 ngày trước

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, while general-purpose proprietary models lack acce…

🅲 hạng C
20
Restoring Diversity in Distilled Image and Video Generators by Probing Activations
📄 Paper HF Daily Papers 7 ngày trước

Distilling diffusion models into few-step student models substantially reduces sampling latency, but induces severe diversity collapse: samples generated from different random seeds show minimal variation. In this work, we investigate the root cause of this failure by probing the…

🅲 hạng C
20
Echo: Coupled Action and Task Status Signal Denoising for Vision Language Action Models
📄 Paper HF Daily Papers 29 ngày trước

Recognizing task completion and failure enables robots to stop upon success and initiate recovery or replanning after failure. Existing methods augment Vision Language Action (VLA) policies with task status signal prediction to support these execution decisions. However, learning…

🅲 hạng C
20
LoCA: Enabling Multi-Modality in Geospatial Location Encoders
📄 Paper HF Daily Papers 28 ngày trước

Location encoders map geographic coordinates to high-dimensional descriptors through contrastive alignment with paired Earth Observation imagery. We prove that this construction is structurally limited: any Lipschitz location encoder produces an output of intrinsic dimension at m…

🅲 hạng C
20
DISC: Domain-Integrated Similarity Cost for Time-Series Dataset
📄 Paper HF Daily Papers 29 ngày trước

The growth of time-series datasets and foundation models has made dataset comparison a central part of time-series workflows. Assessing source-target transfer before model adaptation requires similarity estimates that can summarize transfer-related dataset relationships without e…

🅲 hạng C
20
Speculative Diffusion is a Tale of Two Tradeoffs
📄 Paper HF Daily Papers 29 ngày trước

Speculative decoding is a technique that accelerates the inference speed of autoregressive large language models (LLMs) by sampling from a faster, lower quality draft model. Traditionally, a fundamental characteristic of speculative decoding is that it is lossless with respect to…

🅲 hạng C
20
Unlocking Any-Order Generation in Pretrained Autoregressive Image Models
📄 Paper HF Daily Papers 22 ngày trước

Autoregressive (AR) image generators are powerful in principle but inflexible in practice: training with a fixed raster-scan order hardcodes a single factorization of the image distribution into the model, leaving it unable to inpaint, outpaint, or edit at inference time. Existin…

🅲 hạng C
20
CausalWAM: Learning the Causes of Actions in World Action Models
📄 Paper HF Daily Papers 21 ngày trước

World action models (WAMs) learn latent variables from observations and use these to both predict the evolution of the scene and to generate actions. The prediction has to be of the whole scene, whereas only a portion of the scene causes actions. WAMs fitted to demonstrations do …

🅲 hạng C
20
Cross-Primitive Consistency in Jev
📄 Paper HF Daily Papers 23 ngày trước

Jev enables efficient decision-making with large language models (LLMs) through typed querying primitives, allowing the same semantic state to be expressed in different formats, such as binary and multiple-choice queries. However, the consistency of decisions across primitives ha…

🅲 hạng C
20
Koopman Generative Operators for Efficient Probabilistic Time-Series Forecasting
📄 Paper HF Daily Papers 16 ngày trước

Probabilistic time-series forecasting requires models that simultaneously capture structured temporal dynamics, expressive uncertainty, and efficient inference, yet existing approaches fall short of this goal: latent dynamical models impose structure but rely on restrictive gener…

🅲 hạng C
20
GaussianRenderFormer: Transformer-based Rendering of 3D Gaussian Splats
📄 Paper HF Daily Papers 16 ngày trước

Can a neural network learn to render 3D Gaussian splats on its own, with no rasterizer in the renderer? We present GaussianRenderFormer, a transformer that takes a set of Gaussians and a camera and outputs the image directly. Each Gaussian becomes one token in a view-independent …

🅲 hạng C
20
TaoMate-HD: Progressive-Resolution Autoregressive Audio-Video Generation
📄 Paper HF Daily Papers 18 ngày trước

Streaming digital humans produce detailed video while continuing speech as new text arrives. High-resolution denoising is costly, and text updates can cause repeated or interrupted speech. We present TaoMate-HD, an autoregressive model for high-resolution audio-video streaming. F…

🅲 hạng C
20
Information Games: Strategic Crowding and Firm Repositioning in Language-Model Space
📄 Paper HF Daily Papers 15 ngày trước

When firms respond to new market opportunities, their capacity choices alter the future discounted free cash flows available to both competitors and investors. In competitive markets, these investment decisions publicized in news articles and analyst reports, measuring the change…

🅲 hạng C
20
JoVox: Joint Hierarchical Flow Matching for Lightweight Voxel Generation
📄 Paper HF Daily Papers 12 ngày trước

Generating high-fidelity voxels while keeping their lightweight nature remains an open problem in computer vision. Current generators are either high-fidelity but computationally heavy, or lightweight but of lower fidelity. Hierarchical methods keep the lightweight nature of voxe…

🅲 hạng C
20
onóna: Twin Embeddings with Tied Unembeddings
📄 Paper HF Daily Papers 13 ngày trước

Weight tying reduces the parameter cost of language models, but the usual tied output head is limited to a single softmax. We introduce onóna, a two-table parameterization that increases output expressiveness while preserving input-output weight sharing. The Transformer takes the…

🅲 hạng C
20
Predictive Surprise as Self-Grounding Concept Bottleneck for Interpretable Time Series
📄 Paper HF Daily Papers 13 ngày trước

Time-series classifiers deployed in risk-sensitive domains must be (1) accurate, (2) interpretable, and (3) able to abstain when evidence is weak. Existing methods do not deliver all 3 at once, and the obstacle is structural, rather than logistical. Meaningful temporal patterns c…

🅲 hạng C
20
EVAThink: Evidence Sufficiency-Guided Training for Adaptive Thinking in Video Understanding
📄 Paper HF Daily Papers 10 ngày trước

Under a fixed frame budget, video understanding with vision-language models (VLMs) depends on whether the sampled frames provide sufficient evidence for the question. Sufficient frame evidence supports direct answering, whereas less informative observations may benefit from reaso…

🅲 hạng C