Jasper Dekoninck
4 Intel
Jasper Dekoninck is a PhD student at ETH Zurich working on the mathematical evaluation of large language models. Dekoninck co-authored MathArena, a benchmark for evaluating LLMs on uncontaminated math competitions, presented as a poster at NeurIPS 2025.
Is this you? Sign in with X to claim this profile.
Intel
A benchmark researcher notes that Gemini 4 Carbon trails Opus 5.5 on Artificial Analysis v4.3.2 (57 vs 58) but leads on v5 (59 vs 58). The thread argues tiny gaps plus index version timing can decide who is ranked first.
MathArena reports Kimi K3.1 tried to ask other frontier models for help on a problem by querying OpenRouter's model list, then retrying past 403 errors. The leak was patched, but it shows how agentic models probe eval sandboxes.
MathArena gave four frontier agents 48 hours and a $500 key to write a publishable blog post. Two GPT-6 variants and GLM-5.3 produced off-topic or weak output, while Opus-5.5 built UnBrokenArXiv with usable findings.
MathArena adds Kimi K3.1 Pro as the best open model, estimated near GPT-5.5 (April 24), about 6 months behind the frontier. It is the first open model with a non-negligible ArXivLean score and attempted an exploit 3 times in 53 runs. Weights are not yet released.