Life-Harness evolves reusable interventions from training trajectories to enhance frozen LLM agents on unseen tasks across seven deterministic environments, yielding 88.5% average relative improvement in 116 of 126 model-environment settings.
hub
arXiv preprint arXiv:2405.18369 , year=
11 Pith papers cite this work, alongside 2 external citations. Polarity classification is still indexing.
11
Pith papers citing it
2
external citations · external index
hub tools
citation-role summary
background 3
citation-polarity summary
roles
background 3polarities
background 3representative citing papers
Fast-Slow Training uses context optimization as fast weights alongside parameter updates as slow weights to achieve up to 3x better sample efficiency, higher performance, and less catastrophic forgetting than standard RL in continual LLM learning.
Proposes compiling preference pairs into readable natural-language specifications for inference-time LLM alignment, claiming outperformance over DPO on dense-preference domains.
PACE coordinates low-risk prompt evolution with validated higher-risk control-logic updates to improve frozen SLM agents on benchmarks without model retraining.
NCCE reframes context engineering as instance-level recommendation via bootstrapped anchor contexts and a co-evolving neural collaborative filtering router that assigns specialized contexts per input.
ExecTune trains guide models via acceptance sampling, supervised fine-tuning, and structure-aware RL to boost executability of strategies for black-box LLMs, yielding up to 9.2% higher accuracy and 22.4% lower cost on math and code tasks.
The paper maps agent memory research via three forms (token-level, parametric, latent), three functions (factual, experiential, working), and dynamics of formation/evolution/retrieval, plus benchmarks and future directions.
A rule-aware modular prompt framework enables LLMs to perform structured numeric reasoning on power grid data by separating rules from normalized deviations, improving anomaly detection consistency and reducing token use in IEEE 118-bus tests.
ORFuzz presents the first evolutionary testing framework for LLM over-refusal together with a new benchmark of 1,855 cases that triggers over-refusal at 63.56% average across ten models.
A comprehensive review of self-evolving AI agents that improve themselves over time, organized via a framework of inputs, agent system, environment, and optimizers, with domain-specific and safety discussions.
Self-prompting combined with QLoRA and DPO on small open-weight models yields micro F1 scores up to 0.864 on clinical named entity recognition from 1,200 dental notes.
citing papers explorer
-
Adapting the Interface, Not the Model: Runtime Harness Adaptation for Deterministic LLM Agents
Life-Harness evolves reusable interventions from training trajectories to enhance frozen LLM agents on unseen tasks across seven deterministic environments, yielding 88.5% average relative improvement in 116 of 126 model-environment settings.
-
Learning, Fast and Slow: Towards LLMs That Adapt Continually
Fast-Slow Training uses context optimization as fast weights alongside parameter updates as slow weights to achieve up to 3x better sample efficiency, higher performance, and less catastrophic forgetting than standard RL in continual LLM learning.
-
Towards Spec Learning: Inference-Time Alignment from Preference Pairs
Proposes compiling preference pairs into readable natural-language specifications for inference-time LLM alignment, claiming outperformance over DPO on dense-preference domains.
-
PACE: Two-Timescale Self-Evolution for Small Language Model Agents
PACE coordinates low-risk prompt evolution with validated higher-risk control-logic updates to improve frozen SLM agents on benchmarks without model retraining.
-
Contexting as Recommendation: Evolutionary Collaborative Filtering for Context Engineering
NCCE reframes context engineering as instance-level recommendation via bootstrapped anchor contexts and a co-evolving neural collaborative filtering router that assigns specialized contexts per input.
-
ExecTune: Effective Steering of Black-Box LLMs with Guide Models
ExecTune trains guide models via acceptance sampling, supervised fine-tuning, and structure-aware RL to boost executability of strategies for black-box LLMs, yielding up to 9.2% higher accuracy and 22.4% lower cost on math and code tasks.
-
Memory in the Age of AI Agents
The paper maps agent memory research via three forms (token-level, parametric, latent), three functions (factual, experiential, working), and dynamics of formation/evolution/retrieval, plus benchmarks and future directions.
-
A Rule-Aware Prompt Framework for Structured Numeric Reasoning in Cyber-Physical Systems
A rule-aware modular prompt framework enables LLMs to perform structured numeric reasoning on power grid data by separating rules from normalized deviations, improving anomaly detection consistency and reducing token use in IEEE 118-bus tests.
-
ORFuzz: Fuzzing the "Other Side" of LLM Safety -- Testing Over-Refusal
ORFuzz presents the first evolutionary testing framework for LLM over-refusal together with a new benchmark of 1,855 cases that triggers over-refusal at 63.56% average across ten models.
-
A Comprehensive Survey of Self-Evolving AI Agents: A New Paradigm Bridging Foundation Models and Lifelong Agentic Systems
A comprehensive review of self-evolving AI agents that improve themselves over time, organized via a framework of inputs, agent system, environment, and optimizers, with domain-specific and safety discussions.
-
Self-Prompting Small Language Models for Privacy-Sensitive Clinical Information Extraction
Self-prompting combined with QLoRA and DPO on small open-weight models yields micro F1 scores up to 0.864 on clinical named entity recognition from 1,200 dental notes.