Category
8,685 papers
This work tests whether separating three-state extraction (present, absent or unknown, by an LLM) from decision logic (deterministic code computing score bounds over unknown inputs) lets a system ask only questions that can change the decision.
Names are personal identifiers, but they also carry social meaning and are widely used to evaluate how language models treat different people, so it is shown that this assumption often fails at the lexical interface: matched names are not necessarily matched inputs.
This work presents a diagnostic and optimization framework grounded in a key empirical finding: the value of tokens within a CoT reasoning sequence is highly non-uniform, and this non-uniformity can be effectively characterized by token-level log probability signals.
A single random Gaussian probe gives an unbiased estimate of the squared Frobenius norm of a layer's quantization error. The estimator is well-behaved because round-to-nearest error is spectrally flat. Across 1,683 tensors from a 35B MoE and a 9B dense model, effective dimensionality is 0.93 to 0.96 times the i.i.d. no...
Reach audiences
Advertise in front of researchers, engineers, and readers.
Verifiable QA Generation for Self-Evolving Models (VQS) is presented, which changes how the model judges answers and improves Qwen3-VL by up to 3.18 points at the 2B, 4B, and 8B scales and outperforms the strongest self-evolving baseline at each.
Entropic Advantage Policy Optimization (EAPO) is introduced, an entropy-guided credit assignment method that treats success and failure asymmetrically and promotes more effective exploration, broadening problem coverage and generating more diverse candidate answers.
BOReFT is introduced, which learns a compact, low-dimensional space of hidden-state interventions in a frozen language model, and uses this space as the search domain for Bayesian optimization with an external scoring function, and provides a principled new bridge between discrete proposal spaces of LLM-based search an...
It is suggested that LLMs can support revision boundary judgment when task-adapted, but general purpose prompting alone may not outperform transparent structural heuristics.
A standardized evaluation framework for unstructured NLA explanations is introduced, measuring information recoverable from explanations, contextual support for their claims, and writing quality, and opening up a direction for improving activation-derived training to encourage more informative, supported, and readable...
Context-aware dynamic trees allocate the speculative decoding budget according to draft path probabilities, adapting their depth and branching to the current context. Under stochastic decoding, however, we find that this structural advantage does not always compensate for the acceptance gains of random sampling paired...
This work attempts to make use of high-quality public documents with small perturbations and empirically finds that LLMs can successfully generate context-dependent reasoning traces and answers, which are then used to train a student model that substantially improves the performance on CL-bench.
This work proposes a masked diffusion language model, fine-tuning LLaDA-8B-Instruct with a class-conditional reconstruction objective using LoRA and requiring no architectural changes beyond the base model, which leads on average rank against discriminative baselines trained on stronger backbones across seven held-out...
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.
