Tag
CHIME is a credit-aware hierarchical memory framework that separates planning and execution memory banks to improve long-horizon agentic planning by accurately attributing task outcomes and outperforming baselines.
The paper introduces Recuris, a recursive memory architecture that improves long-horizon agent success by tracking progress and guiding skill selection through localized, validation-gated updates.
This paper proposes SDAM, a memory-based framework for complex Text-to-SQL that uses structure-difference aware reasoning, contradiction-aware reflection, and schema-grounded memory evolution to improve SQL generation. Experiments show modest gains on BIRD-dev and Spider-test benchmarks.
Introduces RSMeM, a knowledge-enhanced memory evolution mechanism for remote sensing agents that bootstraps LLMs with domain knowledge and iteratively integrates failure experience to improve multi-step tool execution, achieving 6% accuracy gain on DeepSeek-V3.2 with minimal additional tokens.
EvoArena introduces a benchmark for evaluating LLM agents in dynamic environments with progressive updates across terminal, software, and social domains, while EvoMem proposes a patch-based memory paradigm that records structured evolution; experiments show current agents achieve only 39.6% accuracy on EvoArena, and EvoMem yields average gains of 1.5% on the benchmark and improvements on GAIA and LoCoMo.
SAGE proposes a novelty gate for memory evolution in agentic LLMs, using a von Mises-Fisher-based density estimator to decide whether to add, merge, or ignore new facts, reducing LLM calls while maintaining memory quality.