Artificial Intelligence

cs.AIComputer Science7 papers

Every entry carries a structured overview and the reproduction evidence collected so far. A paper is listed here when it carries the category as its primary field or as a secondary classification. Both are shown on the paper page.

Latest in cs.AI

Showing 1–7 of 7

Apr 22, 2026Formal Sciences260423.0012

V-tableR1: Process-Supervised Multimodal Table Reasoning with Critic-Guided Policy Optimization

Yubo Jiang, Yitong An, Xin Yang, Abudukelimu Wuerkaixi +6 more

V-tableR1 is a reinforcement learning framework that trains multimodal language models to perform rigorous, step-by-step reasoning on tables rather than relying on pattern matching. It uses a critic model to provide feedback on visual reasoning chains and a novel optimization algorithm (PGPO) to improve performance.

Apr 22, 2026Formal Sciences260423.0014

AAC: Admissible-by-Architecture Differentiable Landmark Compression for ALT

An T. Le, Vien Ngo

AAC introduces a differentiable landmark-selection module for ALT shortest-path heuristics that maintains admissibility by construction through row-stochastic mixtures of triangle-inequality bounds. The method achieves near-optimal landmark coverage on road networks while being 1.2–1.5× faster than classical farthest-point sampling at matched memory.

Apr 22, 2026Formal Sciences260423.0029

pAI/MSc: ML Theory Research with Humans on the Loop

Mahmoud Abdelmoneum, Pierfrancesco Beneventano, Tomaso Poggio

pAI/MSc is an open-source multi-agent system designed to assist researchers in ML theory by automating the conversion of hypotheses into manuscript drafts, significantly reducing manual steering effort. The system emphasizes human-in-the-loop collaboration rather than fully autonomous research.

Apr 20, 2026Formal Sciences260421.0038

MathNet: a Global Multimodal Benchmark for Mathematical Reasoning and Retrieval

Shaden Alshammari, Kevin Wen, Abrar Zainal, Mark Hamilton +4 more

MathNet is a large-scale, multilingual dataset of 30,676 Olympiad-level math problems from 47 countries spanning two decades, designed to benchmark both mathematical reasoning in generative models and mathematical retrieval in embedding systems. The benchmark reveals that even state-of-the-art models struggle with these problems, with top models achieving only 78.4% accuracy, and that retrieval quality significantly impacts retrieval-augmented generation performance.

Apr 20, 2026Formal Sciences260421.0045

Benchmarking System Dynamics AI Assistants: Cloud Versus Local LLMs on CLD Extraction and Discussion

Terry Leitch

This paper benchmarks cloud and local large language models on two System Dynamics tasks: extracting causal loop diagrams and providing interactive coaching. The best local models match mid-tier cloud performance on diagram extraction (77%) but struggle with long-context error-fixing tasks, with backend implementation choices mattering more than quantization levels.

Apr 20, 2026Formal Sciences260421.0064

Using large language models for embodied planning introduces systematic safety risks

Tao Zhang, Kaixian Qu, Zhibin Li, Jiajun Wu +3 more

This paper introduces DESPITE, a benchmark of over 12,000 tasks to evaluate safety in LLM-based robotic planning, revealing that even models with near-perfect planning ability produce dangerous plans 28% of the time. The study shows planning ability scales with model size while safety awareness remains relatively flat, creating a critical gap for deploying LLMs in robotics.

Apr 20, 2026Formal Sciences260421.0074

Learning from Less: Measuring the Effectiveness of RLVR in Low Data and Compute Regimes

Justin Bauer, Thomas Walshe, Derek Pham, Harit Vishwakarma +3 more

This paper investigates how well small language models can learn reasoning tasks through reinforcement learning when training data and compute are limited. The study finds that mixing easy and hard problems during training provides up to 5x better sample efficiency than training on easy problems alone.