저장한 도구도구 추천
FOXAI SEARCH

전체 검색

뉴스 기사 221

연구 동향

INC-MoE: 전문가 병렬 혼합-전문가 추론을 위한 네트워크 내 집단 수렴과 전문가 계산의 중복 처리

arXiv:2610.36339v1 발표 유형: 신규 초록: 전문가 병렬 혼합-전문가(MoE) 서비스에서 디스패치는 토큰을 해당 토큰이 선택한 전문가를 호스팅하는 가속기로 이동시키고, 결합은 각 토큰의 소유자에게 전문가 출력을 다시 모아줍니다. 두 작업 모두 핵심 경로에 위치해 다노드 배포 환경에서 MoE 레이어의 지연 시간을 주도할 수 있습니다. 우리는 INC-MoE를 제안합니다. 이는 디스패치와 결합 작업을 네트워크 스위치로 오프로드하고, 두 집단 수렴을 서로 및 전문가 계산과 겹치도록 설계된 전문가 병렬 MoE 레이어입니다. 디스패치는 스위치 측 멀티캐스트를 통해 네트워크 내 AllGather를 사용하며, 결합은 스위치 측 감축을 통해 네트워크 내 ReduceScatter를 사용합니다. 두 집단 수렴 모두 최종 호스트와 스위치 간의 메시지 세미틱을 활용합니다. 네트워크 내 집단 수렴은 서로 반대 방향으로 로드되기 때문에

arXiv cs.DC
model-evals

Lipi-Ghor: A Large-Scale, Diarized, Multi-Domain Bengali Speech Corpus and a Scalable Pipeline for Building It

arXiv:2610.36184v1 Announce Type: new Abstract: Bengali, spoken by over 230 million people, remains severely under-resourced in spoken language research. Existing open Bengali ASR corpora are confined to short, read or crowd-sourced speech and lack speaker diarization and long-form audio. We present Lipi-Ghor, a large-scale weakly-supervised Bengali speech corpus comprising 885 hours of audio sourced from 998 YouTube videos across 596 unique channels, spanning 17 diverse domains including audiobooks, drama, religious sermons, news, and regional dialects. Each recording is annotated in SSTT format (Speaker, Speech, Transcription, Timestamp), which combines automated diarization with Bengali YouTube caption tracks. To our knowledge, Lipi-Ghor is the first Bengali speech corpus to pair large-

번역 서비스가 일시 중지되어 원문을 표시합니다.
arXiv cs.CL
연구 동향

AK_QMUL at ArGuard Shared Tasks: A Single Fine-Tuned JAIS for Arabic Harmful-Prompt Detection

arXiv:2610.36360v1 Announce Type: new Abstract: We present AK_QMUL's submissions to the ArGuard-2026 shared tasks on Arabic harmful-prompt detection: a binary safe/unsafe task (B1) and a seven-way harm-domain task (B2). Both submissions use one backbone -- JAIS-6.7B fine-tuned with 4-bit QLoRA and a linear classification head over its last-token representation. For B1 we train a four-way head that factors the label by dialect register {safe, unsafe} x {MSA, other}, we augment the training fold with MSA prompts from AraSafe, and decode with a plain argmax to score safe/unsafe. For B2 we fine-tune the same backbone as a seven-way classifier and average the softmax outputs of five random seeds. Both submissions ranked first on their track, with macro-F1 0.984 on B1 and 0.790 on B2. We report

번역 서비스가 일시 중지되어 원문을 표시합니다.
arXiv cs.CL
모델 업데이트

Causal Localization of Evidence Effects in Transformers

arXiv:2610.36524v1 Announce Type: new Abstract: Can controlled symbolic inputs reveal attention heads that influence a language model's decisions from natural language evidence? We introduce controlled evidence aggregation (CEA), which varies the counts and order of two symbols in otherwise fixed prompts and measures changes in a frozen Transformer's preference between two answer tokens. The preference is measured by their logit difference; the prompts prescribe no correct answer. Balanced utility causal selection (BUC) identifies compact head sets by jointly evaluating how much ablating their outputs removes the count effect and how closely patching mean outputs between count conditions recovers the effect in both directions. We derive conditional bounds linking head attribution to interv

번역 서비스가 일시 중지되어 원문을 표시합니다.
arXiv cs.CL
모델 업데이트

Polymathic AI and Princeton Researchers Release FlashMap, a Flow Map Drafter for 6.14x Faster LLM Decoding

FlashMap is a speculative decoding drafter from Polymathic AI, Princeton and USC ISI researchers. It drafts 16 tokens jointly in one pass and reaches up to 6.14x lossless speedup on Qwen3-8B. The post Polymathic AI and Princeton Researchers Release FlashMap, a Flow Map Drafter for 6.14x Faster LLM Decoding appeared first on MarkTechPost.

번역 서비스가 일시 중지되어 원문을 표시합니다.
마크테크포스트
model-evals

Nvidia releases all-in-one adaptable Nemotron 3.5 audio transcription and translation model

Nvidia Corp. today released a leaderboard-winning open artificial intelligence model for multilingual audio transcription and translation, along with an open data pipeline for fine-tuning Nemotron 3.5 Transcribe. Traditional transcription AI models follow a workflow that detects speech, transcribes it, identifies and labels different speakers, then answers questions about the transcription. Transcribe does all of this […] The post Nvidia releases all-in-one adaptable Nemotron 3.5 audio transcription and translation model appeared first on SiliconANGLE.

번역 서비스가 일시 중지되어 원문을 표시합니다.
실리콘앵글 AI
model-evals

Fix evaluation bias to unlock better models on Amazon Bedrock

Property Finder found that 96.8% of their vision-AI evaluation labels were inherited from the incumbent model, hiding its true accuracy. See how they built unbiased ground truth, applied evolutionary prompt optimization, and selected a more accurate, cost-effective model on Amazon Bedrock, cutting cost by 86% across 11 image-quality attributes.

번역 서비스가 일시 중지되어 원문을 표시합니다.
AWS 머신러닝 블로그