cập nhật 1 giờ trước

⏱ Độ trễ thật của tab này: trung bình 6.8 giờ sau khi CVE được công bố (n = 5447, sớm nhất 0.1 giờ). Số âm nghĩa là ta thấy trước ngày công bố.

23
VFlash: Fast, Accurate Visual Planning via Flash Primitive Decoding
📄 Paper HF Daily Papers 14 ngày trước

Models for physical AI tasks, such as robotics and autonomous driving, need not reason solely in language. They can use visual plans--points, bounding boxes, and trajectories--to represent spatial goals and anticipate motion. However, autoregressive vision-language models generat…

🅲 hạng C
23
The novelty exposed: ablation, reproducibility and transparency in 40 technical reports of open-weight large language models (LLMs)
📄 Paper HF Daily Papers 4 ngày trước

In 2025 alone, 48 new large language model (LLM) architectures were added to the Hugging Face transformers library, and many of them arrive with a technical report that presents the model as new. What is actually new in these models, and is it tested? We put this question to LLM …

🅲 hạng C
23
IdeaTaste: Do LLMs Share the Research Taste of Human Researchers?
📄 Paper HF Daily Papers 17 ngày trước

Research taste, the judgment of what to read, which problems to pursue, and which methods to try, increasingly shapes the work of AI scientists, yet it remains far less understood than their ability to execute research. We introduce IdeaTaste, a benchmark that scores an LLM by ho…

🅲 hạng C
23
EvoMemState: Evolving Memory State for Personal Agents
📄 Paper HF Daily Papers 17 ngày trước

Long-term memory enables large language model (LLM) agents to provide continuous personalized assistance by retaining user information across interactions. Existing systems support this process through fact extraction, organization, updating, and retrieval, but largely maintain m…

🅲 hạng C
23
SteerDistill: Quantizing LLM Agents via Closed-Loop Teacher Steering
📄 Paper HF Daily Papers 9 ngày trước

Low-bit quantization reduces the inference cost of LLM agents but can compromise task success. Examining failed rollouts, we find that quantized agents overlook task requirements or environmental conditions, producing environment transitions that diverge from successful full-prec…

🅲 hạng C
23
Continuation-Guided LLM Evolution of Program Policies and Goal Compilers in Craftax
📄 Paper HF Daily Papers 5 ngày trước

Coding agents can synthesize and optimize code-based policies from simulator feedback, allowing the resulting program to act with zero inference model calls. But for long horizon tasks, two challenges arise: local improvements can hurt final success and switching to a new objecti…

🅲 hạng C
23
Quant-Arena: Benchmarking LLM Agents for Quantitative Trading Research
📄 Paper HF Daily Papers 29 ngày trước

Quantitative research is an iterative loop: a researcher proposes a market hypothesis, implements it as an executable signal, tests it on historical data, and decides what to keep. Existing benchmarks score formulas, code, or individual trading decisions, leaving open whether lar…

🅲 hạng C
23
Memory-Centric Continual Learning for AI Agents
📄 Paper HF Daily Papers 29 ngày trước

Autonomous agents powered by foundation models can reason, plan, and use tools, but their fixed weights and context limits hinder sustained learning from experience over long-term deployment. While external memory systems attempt to bridge this gap, they mostly act as passive ret…

🅲 hạng C
23
TraceForge: Distilling with Your Own Thinking
📄 Paper HF Daily Papers 29 ngày trước

Using an LLM to generate training data for another is among the most general ways to train one, and it remains common practice. Some frontier teams have closed that route for their own models: what comes back is an answer, or a short summary, without the detailed reasoning trace …

🅲 hạng C
23
The Surprising Effectiveness of Deleting Weights in LLM Post-Training
📄 Paper HF Daily Papers 12 ngày trước

How little of a pretrained large language model (LLM) has to change for it to acquire new capabilities? We find that zeroing fewer than 0.05% of its weights, with no other modification, matches full fine-tuning (FFT) at scale across three LLM post-training regimes: reinforcement …

🅲 hạng C
23
S3AC: Stabilized, Selective, and Scalable Actor-Critic for LLM Reasoning
📄 Paper HF Daily Papers 24 ngày trước

Critic-free methods are widely used in reinforcement learning with verifiable rewards (RLVR) for long chain-of-thought (CoT) reasoning, avoiding difficult and costly critic training. Common alternatives lack token-level credit assignment and typically rely on grouped rollouts to …

🅲 hạng C
23
SIRIS: Can LLMs Tell When Simulation Errors Matter in AI-Driven Systems Research
📄 Paper HF Daily Papers 13 ngày trước

LLM agents increasingly drive systems research and rely on simulators to evaluate candidate designs at scale. Simulators are imperfect by design: their modeling limitations can distort predictions yet leave the recommended design correct, or change which design should be selected…

🅲 hạng C
23
MotionGraphicsBench: Evaluating Motion Graphics Generation with Programmatic Verification
📄 Paper HF Daily Papers 12 ngày trước

Large language models can generate motion graphics from static images and natural language requests, but it remains difficult to evaluate how well models can follow requests that combine spatial, appearance, and temporal requirements. We introduce MotionGraphicsBench, an expert-c…

🅲 hạng C
23
LoopFill: Pipelined Multi-Token Speculative Decoding for Looped Transformers
📄 Paper HF Daily Papers 5 ngày trước

Looped Transformers improve parameter efficiency by repeatedly applying the same Transformer layers. During autoregressive decoding, however, generating each token requires several sequential passes through these layers, leading to high decoding latency. Pipelined self-speculativ…

🅲 hạng C
23
CulturalRiddles: A Multicultural Benchmark of Riddles
📄 Paper HF Daily Papers 29 ngày trước

As LLMs expand their language coverage, evaluating how well they understand and represent diverse cultures becomes increasingly important. However, many benchmarks remain English-centric, with limited cultural coverage and a focus on factual knowledge over its application in real…

🅲 hạng C
23
On the Origin and Evolution of Training-Inference Mismatch in LLM Reinforcement Learning
📄 Paper HF Daily Papers 21 ngày trước

On-policy reinforcement learning for large language models assumes that training responses are sampled from the current policy. In practice, an inference engine samples the responses and a separate trainer computes the gradients, and the training-inference mismatch between the tw…

🅲 hạng C
23
DecisionRank: On-Policy Distillation for Decision-Model Rerankers
📄 Paper HF Daily Papers 20 ngày trước

Decision models target structured tasks by directly predicting choices or scores, avoiding autoregressive decoding latency. However, transferring a generative LLM's ranking knowledge to a decision model remains challenging: the teacher expresses preferences through a ranked list …

🅲 hạng C
23
FlowTranslator: Training-Free In-Image Machine Translation with Flow-Based Editing
📄 Paper HF Daily Papers 16 ngày trước

In-image machine translation (IIMT) replaces text in an image with its translation while preserving the surrounding visual context. Existing task-specific models achieve convincing performance but require massive paired training images before and after translation. As advanced te…

🅲 hạng C
23
Frontier3DBench: Benchmarking Coding Agents for 3D Creation across Objects, Scenes, and Interactive Open Worlds
📄 Paper HF Daily Papers 13 ngày trước

Coding agents combine visual understanding with program execution, offering a new route to automated 3D creation. By translating images and instructions into editable programs, they bring reconstruction, generation, and editing into a shared workflow. Yet existing benchmarks frag…

🅲 hạng C
23
AutoCompress: Automated Hardware-Guided Model Compression
📄 Paper HF Daily Papers 14 ngày trước

LLM pruning must preserve quality while meeting hardware- and workload-specific constraints on latency, throughput, and memory. Existing pruning workflows often target parameter counts, requiring practitioners to manually select model sizes that meet latency, throughput, and GPU …

🅲 hạng C
23
When Is Proof-Step Verification Worth Its Cost? Operating Requirements of Certify-or-Decline Verification for Open-Weight Mathematical Reasoning
📄 Paper arXiv cs.AI 6 ngày trước

arXiv:2611.14794v1 Announce Type: new Abstract: Proof-step verification rewrites a solver's answer into typed steps and uses LLM judges to certify or decline. We evaluate this pipeline on open-weight models across GPQA, AIME-2025, and a MATH-500 subset. Certification raises answ…

🅲 hạng C
23
AutoAssumption: Searching over Assumptions Discovers ML-Systems Algorithms
📄 Paper arXiv cs.AI 6 ngày trước

arXiv:2611.14844v1 Announce Type: new Abstract: Many advances in ML systems came from breaking an assumption that earlier methods took for granted: speculative decoding dropped the assumption that a model needs one forward pass per token. Yet algorithms rarely state their assump…

🅲 hạng C
23
A physics-constrained deep reinforcement learning method for autonomous mission planning of Titan deep space probes
📄 Paper arXiv cs.AI 6 ngày trước

arXiv:2611.15389v1 Announce Type: new Abstract: Titan is a compelling target for future deep space exploration, while its extreme Earth-Titan distance and complex environment make continuous ground-in-the-loop planning impractical. Existing planning methods, primarily developed …

🅲 hạng C
23
VRMaaS: Verifiable Reward-Model-as-a-Service for Trustworthy LLM Alignment
📄 Paper arXiv cs.CR 6 ngày trước

arXiv:2611.14984v1 Announce Type: new Abstract: Reward models (RMs) are increasingly deployed as service components in LLM alignment pipelines, where payment, audit, moderation, and governance workflows act on the scores they return. An inference proof alone does not show whethe…

🅲 hạng C