Category
9,568 papers
Redline is a finite-sample procedure that selects operating points, hand-picked or learned, from the correctness of their answers on calibration prompts, and keeps the reference-relative risk, the joint probability that the reference answers correctly and a candidate configuration does not, within a user-chosen budget...
Autoregressive text-to-speech (TTS) systems synthesize natural speech but, once trained, offer little control over speaking rate. We show that speaking rate can be steered at inference time, without retraining, by clamping a single decoder block's activation along a discovered speed axis. A decoder-block analysis recov...
Redwing is proposed, a design principle for robust token-level watermarking that generalizes to TTS models at a speech-quality cost close to that of KGW and shows that retokenization is not merely a source of noise: its transition structure can be exploited as a design principle for robust token-level watermarking.
This work identifies what must be preserved beyond the replaced value when using a KV cache as updatable memory and recovers distinctions missed by fixed cache similarities on synthetic record pairs.
Reach audiences
Advertise in front of researchers, engineers, and readers.
This work identifies three node categories (common temporal, common to the year, and chrono-semantic) which interact to generate a temporal filter during factual recall and provides the first feature-level map of temporal recall by isolating individual MLP features via transcoder circuit tracing.
CAME (Company-Aware Evidence-Memory Experts), a residual-forecasting framework that refines a no-leakage statistical anchor when current semantic evidence and prior error patterns justify an adjustment, is introduced.
What enters an LLM's context can influence its decision even when it should not determine it, and exposure systematically shifts decisions, with effects reaching nearly 50 percentage points in closed-weight models.
Across twelve models, dense and hybrid, the authors move a hidden state from one story into another where the two routes lead to different places, and read off which place the model gives, which makes the answer a handoff.
The USAI-Quant benchmark is developed, the first benchmark designed to quantitatively evaluate VLM's reasoning capabilities on built environment metrics via remote sensing imagery, and reveals that current state-of-the-art models consistently fall short on numeric reasoning tasks.
OpenTumorBoard is introduced, a benchmark with 611 patient cases and 19,157 discussion turns across ten specialist roles, transcribed from 12,534 minutes of publicly available tumor board recordings on YouTube, and automated curation pipeline is released to support the development and evaluation of LLMs for multidiscip...
KV-Lingo can be used for seamless model switching, staying close to re-prefill across repeated switches in the authors' multi-turn evaluations, and is found particularly useful for dynamic model routing.
HERO-MoE, Historical Expert ROuting with Scale-Preserving Fusion, a routing framework that injects historical routing priors into MoE routers by reusing detached, dense routing distributions collected from preceding MoE layers by reusing detached, dense routing distributions collected from preceding MoE layers.
Natural Language Processing
Our most capable vision-language-action model enhances robot dexterity, reliability, and safety
A Blog post by Ai2 on Hugging Face
A Blog post by IBM Research on Hugging Face
Flint-interactive gives AI agents a language for creating interactive visualizations. Click a chart, filter data, highlight related content, and explore beyond the view the author originally chose. The post Flint-interactive: A language for creating interactive data visualizations appeared first on Microsoft Research.
We’re on a journey to advance and democratize artificial intelligence through open source and open science.
We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.
