3 papers
cs.LG2025
Learning to Undo: Rollback-Augmented Reinforcement Learning with Reversibility Signals
Andrejs Sorstkins, Omer Tariq, Muhammad Bilal
This paper proposes a reversible learning framework to improve the robustness and efficiency of value based Reinforcement Learning agents, addressing vulnerability to value overest…
cs.AI2025
Diagnostics of cognitive failures in multi-agent expert systems using dynamic evaluation protocols and subsequent mutation of the processing context
Andrejs Sorstkins, Josh Bailey, Dr Alistair Baron
The rapid evolution of neural architectures - from multilayer perceptrons to large-scale Transformer-based models - has enabled language models (LLMs) to exhibit emergent agentic b…
cs.CL2025
Assessing RAG and HyDE on 1B vs. 4B-Parameter Gemma LLMs for Personal Assistants Integretion
Andrejs Sorstkins
Resource efficiency is a critical barrier to deploying large language models (LLMs) in edge and privacy-sensitive applications. This study evaluates the efficacy of two augmentatio…