Alessandro Sordoni

The Markovian Thinker: Architecture-Agnostic Linear Scaling of Reasoning

Gistify: Codebase-Level Understanding via Runtime Execution

Effect of Document Packing on the Latent Multi-Hop Reasoning Capabilities of Large Language Models

BugPilot: Complex Bug Generation for Efficient Training of SWE Agents

Training Plug-and-Play Knowledge Modules with Deep Context Distillation

Learning to Solve Complex Problems via Dataset Decomposition

VinePPO: Refining Credit Assignment in RL Training of LLMs

Exploring Sparse Adapters for Scalable Merging of Parameter Efficient Experts

Putting the Value Back in RL: Better Test-Time Scaling by Unifying LLM Reasoners With Verifiers

Not All LLM Reasoners Are Created Equal

VinePPO: Unlocking RL Potential For LLM Reasoning Through Refined Credit Assignment