A field on ReadSearch
Computer Science & AI
Much of the new work in computer science and AI appears first as a preprint on arXiv. ReadSearch reads those listings every day, with papers OpenAlex files under the same subjects, and this page shows the newest it has, from machine learning and language to systems, theory and security.
The newest 1,000 papers, from 4 November 2026 to 5 November 2026: 1,000 arXiv preprints.
Newest papers, page 27
A Hidden Markov Preference Model for Online Prediction of Adversary Actions
A defender anticipating an attacker needs a probability distribution over the attacker's next action from information available before that action occurs.VFlash: Fast, Accurate Visual Planning via Flash Primitive Decoding
Models for physical AI tasks, such as robotics and autonomous driving, need not reason solely in language.ReproBridge: Building Reproducibility into ML Research Artifacts
Reproducing machine learning experiments requires reconstructing procedures from information scattered across papers, released code, and supporting artifacts.Unpacking the Compute Efficiency of Strong-to-Weak OPD against GRPO
On-policy distillation (OPD) trains a student on its own rollouts under a frozen teacher's per-token supervision.Tusnády's problem and online two-sample thinning
We determine the sharp order of discrepancy in Tusnády's problem in every fixed dimension $d\ge 2$.Cognitive Tours: Making Understanding a Deliverable in AI-Assisted Development
Generative AI and agentic coding are accelerating software development, leading to concerns about technical debt.Adversarial Training with Large Step Sizes: Implicit Bias, Optimization, and Sharpness
Adversarial Training (AT) is an important method for improving model robustness.Latent RL Priors for Data-Efficient Dexterous Manipulation
Learning dexterous manipulation from limited visual demonstrations is challenging because policies must acquire both perceptual grounding and precise control, while reinforcement learning (RL) in…OopsieRecover: Learning Transferable VLA Corrections from Semantic Interventions
Vision-Language-Action (VLA) models can perform a wide range of robotic manipulation tasks, but often fail during deployment.Towards interpreting any residual stream direction by scaling activation inversion
To interpret a direction in a language model's residual stream, we can look for text that strongly activates it.Canonical Cut Hierarchy in Almost-Linear Time via Reduction to Max-Flow
The canonical cut hierarchy recursively partitions a graph along Tutte-Nash-Williams partitions, choosing the partition with the most number of parts at each step.Same Behavior, Different Mechanism: What Finite Experiments Can Certify About Integrated Information
Interpretability, AI auditing and assessments of machine consciousness increasingly rest on mechanism-level quantities: functionals of a system's causal transition law that describe how it computes…XStack: A Benchmark for ML Repo Migration
ML code should run where researchers choose, not where it was first written.EgoGeneralist: Unifying Egocentric Action-Video Tasks through Video-Native Modeling
Modern video models possess strong visual priors, yet do not natively represent human action in metric 3D space.Scientist World: Grow a Research World for AI-Scientists
AI-Scientists arrived, but a world for them to grow in did not.SwarmEvolve: Latent Agent Swarms for Evolutionary Discovery and Optimization
LLM-guided evolutionary search uses AI agents to improve programs for mathematical and computational problems by iteratively refining solutions and combining high-scoring methods.Where and Why is Grounded Reasoning Useful?
Grounded reasoning, where a model outputs text coordinates of relevant locations in chain-of-thought reasoning, is a common strategy for solving complex visual reasoning problems.Do You Miss the Nights We Spent Crafting Figures? Rethinking Teaser and Framework Figures in the Age of Generative AI
Scientific figures give research ideas a visual form through choices of abstraction, composition, and emphasis.User Embeddings are Superpositions of Interpretable Behavioral Modes
User embeddings are central components of modern recommendation, personalization, and behavioral modeling systems.BobFlow: Towards Human-Agent Collaborative Data Science Using Dataflows
Recent advances in large language models (LLMs) have established human-agent collaboration as a promising paradigm for data science.
Get the few papers worth your time in Computer Science & AI every morning.
Choose your topics and ReadSearch reads the new papers in them each day, then brings you the few worth reading, each with its key finding in a sentence.