This work introduces TCS-Bench, a benchmark for evaluating Large Language Models (LLMs) on research-level Theoretical Computer Science (TCS) proof generation, and benchmarks the verifier against human-expert proof judgements on a set of target statements and generated proofs pairs.
Abstract
We introduce TCS-Bench, a benchmark for evaluating Large Language Models (LLMs) on research-level Theoretical Computer Science (TCS) proof generation. TCS-Bench consists of theorem-proving tasks from papers published at top theoretical computer science venues (STOC, FOCS, and SODA). Each task provides the necessary context to derive a self-contained proof for a target result. We evaluate state-of-the-art models on this benchmark. We verify the correctness of generated proofs via a verification agent, and further benchmark the verifier against human-expert proof judgements on a set of target statements and generated proofs pairs. Our reference verifier achieves over 90% accuracy on the expert labeled set.
An automated TCS research framework that generates, formalizes, filters, and proves new claims, and further develops an automated TCS research framework that generates, formalizes, filters, and proves new claims.
Dingzirui Wang, Xuan-Liang Zhang, Keyan Xu et al.· 2 citations
A proof checker called Nörgler is introduced that builds upon and extends the established approach pioneered by GDV and supports checking propositional, (untyped and typed) first-order, and higher-order refutations represented in TSTP.
Melanie Taprogge, H. Sariyanto, Alexander Steen· 1 citation
This is the first, to the authors' knowledge, integration of an LLM with a termination-analysis algorithm, solving 39% more problems than the best LLM alone and more than tripling the number solved by the built-in analysis on their benchmark.
Zeke Medley, P. Manolios· International Conference on...· 0 citations
Evaluating state-of-the-art open LLMs reveals a significant robustness gap, and shows that reliable process-level verification remains challenging, and evaluator robustness should be measured separately from solver accuracy, even in a simple algebraic domain with exact ground truth.
Schedulability analysis is essential for certifying real-time systems, but existing tests are often developed through pen-and-paper proofs that are difficult to scale, validate, and maintain. Mechanized verification in PROSA/ROCQ offers a rigorous alternative, yet manually constructing such proofs requires substantial...
Sadat Shahriyar, Shareef Ahmed, Abdullah Al Arafat· 0 citations
Large language models (LLMs) have achieved remarkable performance on high-school and olympiad-style mathematics, yet their capabilities on advanced mathematics remain poorly understood. Existing benchmarks, however, fall short in both scope and evaluation granularity: they provide limited disciplinary coverage and ofte...
A new computational approach reveals how cells communicate to assemble into tissues. This could accelerate stem cell engineering, improve disease models, and inform therapies.
Microsoft Research Blog· microsoft.comOct 15, 2026
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.
MIT News · Artificial Intelligence· news.mit.eduSep 24, 2026
