AI Networking Cookbook: Practical recipes for AI-assisted network automation and development
Category
9,568 papers
Reward models (RMs) are a key component of reinforcement learning from human feedback, enabling the alignment of large language models with human values such as helpfulness and harmlessness. However, these alignment goals can conflict, leading to interference when optimised jointly. We refer to this conflict as alignme...
Latent chain-of-thought compresses reasoning through continuous hidden-state recurrence, but existing approaches are difficult to optimize with standard on-policy reinforcement learning (RL) and hard to analyze mechanistically. We introduce SWITCH, a switchable latent reasoning framework that addresses both challenges...
The rapid advancement toward long-context reasoning and multi-modal intelligence has made the memory footprint of the Key-Value (KV) cache a dominant memory bottleneck for efficient deployment. While the established per-channel quantization effectively accommodates intrinsic channel-wise outliers in Key tensors, its ef...
Large language model agents increasingly rely on external skills to solve complex tasks, where skills act as modular units that extend their capabilities beyond what parametric memory alone supports. Existing skill-based RL methods assume external skills either accumulate as persistent guidance or internalized into the...
Reach audiences
Advertise in front of researchers, engineers, and readers.
A distilled student can copy its teacher's mistakes along with its skills. On-policy distillation (OPD), used in recent post-training recipes, has the teacher grade every token of the student's own outputs. Which teacher errors does it pass on? We build ErrOPD, a testbed whose teachers are fine-tuned to close a JSON li...
This work proposes TopoCurate, an interaction-aware framework that projects multi-trial rollouts from the same task into a unified semantic quotient topology that explicitly captures how tool invocations and environmental responses drive the divergence between effective strategies and failure modes.
Under GRPO training, the hardened judge eliminates the reward-collapse failures (false-positive spikes and length collapse) the authors observe in the unhardened baseline on both MATH and GSM8K at ten seeds per condition.
Upperological Signatures is presented for characterizing the structure of embedding spaces and finds that it outperforms individual metrics in downstream performance prediction and reveals novel insights into representational similarity that challenge the global claims of the Platonic Representation Hypothesis.
Do multilingual language models learn a universal internal geometry for color, and is it perceptually grounded or primarily linguistic? We decompose color representations into two geometric axes: (P1) alignment to perceptual color space, and (P2) cross-lingual symmetry. Analyzing layerwise embeddings across models and...
Dimensionality reduction can preserve the average performance of text embeddings even when a substantial fraction of their dimensions is removed. Aggregate metrics, however, do not reveal whether degradation is distributed uniformly across examples. We study per-example neighborhood stability under PCA truncation acros...
Longer contexts allow large language models (LLMs) to condition their predictions on more information, but inference costs grow prohibitively with context length. An emerging alternative uses test-time updates to encode context as temporary changes to model weights. However, encoding this information into weights that...
Soft tokens have so far been used to replace discrete chain-of-thought in latent reasoning methods. These approaches typically require multi-stage training and, in some cases, auxiliary modules, yet they still underperform supervised fine-tuning (SFT) ontriples. In this paper, we ask whether soft tokens can be used to...
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.
