Back to Arena

MoT

by Fudan University (Li & Qiu)

System Card

OrganizationFudan University (Li & Qiu)
Released2023-05
Architectureagentic-workflow / Pre-thought externalized memory
DetailsTwo-stage framework: pre-test stage LLM pre-thinks on unlabeled data and saves high-confidence thoughts as external memory; at test time, recalls relevant memory to assist reasoning. No labeled data or parameter updates.
Parameters—
Domainagent-memorylifelong-learning
Open SourceYes
PaperView Paper
CodeRepository
self-improvementemnlp-2023cotpre-thinking

Capability Profile

Benchmark Scores

6 of 14 benchmarks
Long-Context Retrieval
0/5
RULER
no data
NIAH
no data
LooGLE
no data
LongBench
no data
∞Bench
no data
Multi-Turn Recall
2/2
LoCoMo
75.349p
MemoryBank
73.130p
Cross-Session Memory
1/1
LongMemEval
81.588p
Multi-Hop QA
2/3
BABILong
77.975p
MultiHop-RAG
no data
HotpotQA
74.969p
Agent Task Memory
1/1
AgentBench-Mem
7226p
Personalization
0/1
PerLTQA
no data
Factuality / Grounding
0/1
RAGAS
no data
Sources:MoT paper (arXiv:2305.05181); evaluated on LoCoMo: Long-Term Conversational Memory Benchmark (Snap Research, 2402)MoT paper (arXiv:2305.05181); evaluated on LongMemEval: Benchmarking Chat Assistants on Long-Term Interactive Memory (Salesforce AI Research, 2410)MoT paper (arXiv:2305.05181); evaluated on AgentBench Memory Track (Tsinghua KEG, 2308)MoT paper (arXiv:2305.05181); evaluated on BABILong: Testing the Limits of LLMs with Long-Context Reasoning-in-a-Haystack (AIRI, 2406)MoT paper (arXiv:2305.05181); evaluated on HotpotQA: A Dataset for Diverse, Explainable Multi-hop Question Answering (Stanford / CMU, 1809)MoT paper (arXiv:2305.05181); evaluated on MemoryBank: Enhancing LLMs with Long-Term Memory (Sun Yat-sen University, 2305)