논문해외HF Daily Papers
BEND 2.0: Benchmarking the Capabilities Needed to Design DNA
Genomic foundation models are increasingly used to propose DNA sequences and edits.
실시간
국내외 뉴스 · 기업 공식 발표 · 논문 · 영상 · 커뮤니티 소식을 한국시간 기준으로 제공합니다. 같은 소식은 하나로 묶고, 기본은 핵심만 보여 드립니다.
모든 수집 소식(구글 뉴스로 들어온 단건 보도 포함).
Genomic foundation models are increasingly used to propose DNA sequences and edits.
Multimodal large language models (MLLMs) can generate detailed video captions, yet they may overemphasize minor details while overlooking essential context and…
Computer programmers chain together the same reusable computational primitives, for example, conditional statements or loops, but in different orders to…
Attention remains a substantial computational cost in Transformer training, particularly at long context lengths.
Protein-complex structure prediction with AlphaFold 3 depends on MSA content, depth, and cross-chain pairing.
Rapid advances in cowork agents are shifting human-AI collaboration toward real-world automation, delivering software, designs, and other artifacts.
Science experiment videos demonstrate how experiments are performed, but offer no direct interaction for learners.
Classical structure-from-motion reconstruction proceeds iteratively, refining camera poses and scene geometry from a current estimate.
Few-step autoregressive video diffusion enables efficient generation, yet maintaining consistency over long rollouts remains challenging, especially when…
Human-Centred, Risk-Constrained Evaluation of Financial Document Recognition: Evidence, Critical Errors, and Selective Automation
기존 문자·단어 오류율 평가 대신 금융 사실 보존 여부를 반영하는 인간 중심의 위험 제약 평가 프레임워크를 개발했다.
Recent advances in Vision-Language-Action (VLA) models have enabled fine manipulation capabilities, yet performing tasks that require working memory remains a…
Progress in AI is often measured by what a single model can do on its own.
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…
Tool-using language-model agents often execute actions that their domain policies forbid.
AI-generated videos often depict people within detailed scenes while leaving their small facial regions poorly resolved.
Traditional image cropping assumes the photograph already contains the composition worth keeping, and asks only where to crop.
Recently, harness self-evolution has emerged as an important paradigm for improving agent performance.
General-purpose robot policies must work across robots, tasks, and environments, yet the variation that makes physical trials informative also makes them…
Foundation models in changing environments can encounter unfamiliar domain knowledge that their original training did not cover.
Speculative decoding accelerates memory-bound LLM decoding with a lightweight drafter, and parallel drafters extend this speedup by drafting a whole block in…
Long-form multi-shot video editing requires an intended edit to remain coherent across abrupt changes in scene, viewpoint, and subject appearance, while…
3D generation, editing, and completion are typically treated as regenerating the complete 3D asset under task-specific constraints, even when only a local…
Natural Language Processing (NLP) has been widely applied in financial domains, enabling automated analysis and retrieval of stock market information.
GUI web agents require realistic and interactive environments that remain stable during both evaluation and training.
**Purpose:** Ultrasound biometry is widely used across various clinical applications, including prenatal and cardiac examinations, to obtain quantitative…
We introduce Dory1.0-4B, a 4-billion-parameter looped (latent-recurrent) language model with elastic depth: a single checkpoint serves different inference…
Speculative decoding (SD) accelerates large language model inference by reducing the sequential cost of token-by-token generation.
Most image-based 3D reconstructors primarily recover visible surfaces, producing partial point clouds with large holes and leaving unobserved geometry…
Video dubbing aims to synthesize lip movements that match target speech while preserving the speaker identity, head pose, and background of a source video.
LLM agents for GPU kernel optimization sit at two ends of a structure-freedom tradeoff.
Reasoning substantially improves language models' ability to solve difficult problems, but long reasoning traces make these gains expensive.
Streaming digital humans produce detailed video while continuing speech as new text arrives.
Generative AI is billed per token, yet tokens are not equally expensive.
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…
Long-term memory enables large language model (LLM) agents to provide continuous personalized assistance by retaining user information across interactions.
TRIO: Task Refinement with Individual Optimization for Retrieval-Augmented Generation
부적절한 검색과 환각 문제를 개선하기 위해 개별 최적화를 통한 검색 증강 생성(RAG) 성능 향상 방안을 연구한다.
Stochastic Deep Supervision for Gallery-Consistent Early-Exit Face Recognition
SD-ViT는 비전 트랜스포머의 블록 깊이마다 식별 가능한 특징을 출력하도록 학습하여 자원 제약 환경에서도 얼굴 인식이 가능하게 한다.
DepthGraft: Structural Regularization through Hierarchical Cross-Layer KV Reconstruction
대형 언어 모델의 깊이가 깊어짐에 따라 발생하는 정보 희석 문제를 완화하기 위해 교차 레이어 KV 재구성 구조 정규화 기술을 제안한다.