Back
Sydney, Australia 07–12 December
Ryquo Labs × App-In Club at
NeurIPS 2026
We study how AI systems verify decisions, adapt to change, and use evidence. Our work spans agent reliability, human–AI interaction, pathology models, and AI education.
Accepted work
Read paper Language Agent Tree Search: Teaching Agents to Change Their Minds
Jump Start Your Policy Learning with Lessons from 145,000 Training Runs
TwinCheck: Evidence-Grounded Negative-Twin Verification for Stateful Tool Agents
One-Slide Calibration of Pathology Foundation Models
SayWhen: Set-Valued Semantic Commitment for Low-Regret Early Speech
LiveLoRA: Exact KV Cache Migration Across LoRA Updates for LLM Agents
Wrong the Same Way: Cross-Prompt Sharing Both Suppresses and Amplifies Reward Hacking in RLVR
Memory-Efficient Speculative Decoding with Quantized Draft KV Cache
HARMONIQ: A QSAR-Guided Reinforcement Learning Framework for the Design of Small-Molecule MAO-B Inhibitors Targeting Alzheimer’s Disease
Neural Network’s Neurosurgery
December 7–12, 2026