Category
8,685 papers
LLM post-training combines supervised fine-tuning (SFT), a mode-covering forward-KL objective, with reinforcement learning (RL), a mode-seeking reverse-KL objective, with reinforcement learning (RL), a mode-seeking reverse-KL objective to exploit anisotropy, a division of labor that post-training can exploit.
The memory usage and decoding latency of LLM inference grow rapidly with context length. To reduce these costs, key-value (KV) cache compression methods selectively retain cached states based on token importance or differences in attention patterns across heads. However, we discover that retrieval capability varies sub...
Large language models can generate several records within one autoregressive completion, making earlier answers available as context for later records. This paper shows that such shared-completion batching can distort relationships among variables in the resulting synthetic data, using controlled tests on synthetic sur...
HM-Router is introduced, a routing method that jointly selects a model and harness for each query, with an interaction term inspired by canonical polyadic tensor decomposition to capture how their compatibility varies with the query.
Reach audiences
Advertise in front of researchers, engineers, and readers.
Safety auditing under compression should rely on several judges matched to the serving context rather than on keyword rates alone, because the gap is not universal and weakens under short fillers and paired SnapKV, so safety auditing should rely on several judges matched to the serving context rather than on keyword ra...
DoteBench is introduced, a bilingual evaluation suite that measures precise instruction following, local preservation, and audio quality across the four controls and their composition, while audio quality remains comparable to existing open-source systems.
The findings support evaluation practices that separate preference from safety, report safety-critical failure rates directly, and incorporate clinically grounded adjustments when ranking LLMs for clinical decision making.
Despite the growing use of large language models (LLMs) as creativity evaluators, evidence of their alignment with human evaluations remains mixed, raising the question of when and why their judgments converge with or diverge from those of humans. Across three studies and six widely used LLMs, we addressed this gap by...
It is demonstrated that when producing deceptive rather than honest responses, models exhibit a gap between their belief and their commitment, which shows how confidence can be a practical tool for detecting and diagnosing deceptive behaviour in LLMs.
This paper introduces tool efficiency, a new quantitative metric to evaluate the rate of useful tool calls in an LLM agent trajectory, and determines the sign of marginal tool utility for each tool call in a trajectory using LLM-as-a-Judge.
LACUNA is introduced: the first unlearning testbed with ground-truth parameter-level localization and shows that when localization is successful, even a simple gradient-based unlearning method achieves strong erasure and robustness to resurfacing attacks, highlighting the importance of precise unlearning.
GDP-RAG is proposed, a plan-based framework that targets only the information delta based on three simple design choices, and achieves the highest accuracy among all compared systems while maintaining a cost-of-pass of 0.51.
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.
A new computational approach reveals how cells communicate to assemble into tissues. This could accelerate stem cell engineering, improve disease models, and inform therapies.
Zero-knowledge proofs can verify computing results without rerunning workloads or exposing sensitive data. Magnetar helps bring them into the mainstream with support for standard programming languages, major speed gains, and quantum-resistant security. The post Proof-based general-purpose computing with Magnetar appeared first on Microsoft Research.
A new language-processing tool could help identify the highest-risk individuals from natural language, enabling swifter interventions.
A Blog post by Liquid AI on Hugging Face
We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.
