Matrix iteratively refines an LLM agent's memory from training trajectories and outperforms prompting and vanilla agent baselines on private UBL invoice transport-reference extraction, though gains on the released anonymized subset are small.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CL 1years
2024 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Memory-Augmented Agent Training for Business Document Understanding
Matrix iteratively refines an LLM agent's memory from training trajectories and outperforms prompting and vanilla agent baselines on private UBL invoice transport-reference extraction, though gains on the released anonymized subset are small.