3 papers
cs.AI2026
CivBench: A Long-Horizon Benchmark for Tool-Mediated Agents in Civilization VI
Austin Tudor David Andrews, Liam Wilkinson, Jamie Heagerty +3
We present CivBench, an open-source benchmark for evaluating language model agents in long-horizon, tool-mediated environments through the Model Context Protocol (MCP). A single ep…
cs.AI2026
Reason to Play: Behavioral and Brain Alignment Between Frontier LRMs and Human Game Learners
Botos Csaba, Sreejan Kumar, Austin Tudor David Andrews +6
Humans rapidly learn abstract knowledge when encountering novel environments and flexibly deploy this knowledge to guide efficient and intelligent action. Can modern AI systems lea…
cs.LG2024
BP(λ): Online Learning via Synthetic Gradients
Joseph Pemberton, Rui Ponte Costa
Training recurrent neural networks typically relies on backpropagation through time (BPTT). BPTT depends on forward and backward passes to be completed, rendering the network locke…