Tag
A developer tested how small edge models (LFM2.5, Gemma variants) retain a single fact across conversation turns, finding that models often confidently deny knowing information that remains in context, posing a trust issue for agent architectures and suggesting a trade-off between memory and format discipline.
Explores a common failure mode in recursive language models (RLMs) where free-text subagent responses cause issues, and presents a solution using structured outputs to improve reliability, illustrated with a long-context question-answering example from NarrativeQA.
The author describes inverting the textbook agent memory design from retrieval-on-demand to injection-first to avoid latency and confident empty-context errors, detailing the architecture and a dangerous self-poisoning failure mode with write-back.
This paper identifies a novel failure mode in reasoning models called unfaithful capitulation, where the chain-of-thought remains factually correct across adversarial multi-turn dialogues but the final answer flips wrong, highlighting limitations of current evaluation methods.