EXG is an experience graph framework for self-evolving LLM agents that supports online real-time growth and offline reuse to enhance solution quality and efficiency on code generation and reasoning benchmarks.
Canonical reference
In: Proceedings of the 36th Annual ACM Symposium on User Interface Software and Technology (UIST ’23), pp
Canonical reference. 100% of citing Pith papers cite this work as background.
12
Pith papers citing it
Background
100% of classified citations
citation-role summary
background 6
citation-polarity summary
years
2026 12roles
background 6polarities
background 6representative citing papers
Literate execution treats documentation and visualizations as dynamic, computable parts of program execution via provenance tracking, inverting traditional literate programming to make programs more explorable.
ConsumerSim reconstructs official CCI series from synthetic populations and multi-source signals, outperforming baselines on reconstruction metrics and aiding short-horizon activity predictions.
APG4RecSim automatically generates realistic user profiles for LLM-based recommendation simulations, outperforming manual baselines by up to 7% in nDCG@10 and 8% in JSD on three benchmark datasets.
A think-aloud study reveals that AI tools in early research misrepresent uncertainty, obscure provenance, and create fragile trust, leading researchers to develop compensatory strategies to preserve scholarly judgment.
MAESTRO adds a shared preference memory plus GUI-adaptation and workflow-navigation mechanisms to conversational agents with GUIs and tests them in a 33-person movie-booking study.
An LLM acting as real-time controller for SIMP topology optimization parameters outperforms fixed schedules and heuristics, delivering 5.7-18.1% lower compliance on 2D and 3D benchmarks.
LLM assistance shortens idea-generation periods and reduces creative moments during programming tasks while yielding solutions with comparable idea counts and greater functional correctness.
Embodied LLM agents exhibit emergent collaborative behaviors indicating mental models of partners in a color-matching game, detected via LLM judges and supported by positive user feedback.
PSI uses a shared personal-context bus to publish state and write-back affordances, turning isolated AI-generated modules into synchronized, chat-accessible instruments.
Societal-scale LLM agent simulations for policy need three preconditions: avoid neutral treatment of marginalized population simulations, require population participation, ensure accountability, plus development and deployment reports.
citing papers explorer
-
EXG: Self-Evolving Agents with Experience Graphs
EXG is an experience graph framework for self-evolving LLM agents that supports online real-time growth and offline reuse to enhance solution quality and efficiency on code generation and reasoning benchmarks.
-
Literate Execution
Literate execution treats documentation and visualizations as dynamic, computable parts of program execution via provenance tracking, inverting traditional literate programming to make programs more explorable.
-
Uncovering Salience-Driven Dynamics in Consumer Confidence with Generative Social Simulation
ConsumerSim reconstructs official CCI series from synthetic populations and multi-source signals, outperforming baselines on reconstruction metrics and aiding short-horizon activity predictions.
-
Task-Aware Automated User Profile Generation for Recommendation Simulation Using Large Language Models
APG4RecSim automatically generates realistic user profiles for LLM-based recommendation simulations, outperforming manual baselines by up to 7% in nDCG@10 and 8% in JSD on three benchmark datasets.
-
How Researchers Navigate Accountability, Transparency, and Trust When Using AI Tools in Early-Stage Research: A Think-Aloud Study
A think-aloud study reveals that AI tools in early research misrepresent uncertainty, obscure provenance, and create fragile trust, leading researchers to develop compensatory strategies to preserve scholarly judgment.
-
MAESTRO: Adapting GUIs and Guiding Navigation with User Preferences in Conversational Agents with GUIs
MAESTRO adds a shared preference memory plus GUI-adaptation and workflow-navigation mechanisms to conversational agents with GUIs and tests them in a 33-person movie-booking study.
-
Large Language Models as Optimization Controllers: Adaptive Continuation for SIMP Topology Optimization
An LLM acting as real-time controller for SIMP topology optimization parameters outperforms fixed schedules and heuristics, delivering 5.7-18.1% lower compliance on 2D and 3D benchmarks.
-
"Like Taking the Path of Least Resistance": Exploring the Impact of LLM Interaction on the Creative Process of Programming
LLM assistance shortens idea-generation periods and reduces creative moments during programming tasks while yielding solutions with comparable idea counts and greater functional correctness.
-
Evaluating Generative Models as Interactive Emergent Representations of Human-Like Collaborative Behavior
Embodied LLM agents exhibit emergent collaborative behaviors indicating mental models of partners in a color-matching game, detected via LLM judges and supported by positive user feedback.
-
PSI: Shared State as the Missing Layer for Coherent AI-Generated Instruments in Personal AI Agents
PSI uses a shared personal-context bus to publish state and write-back affordances, turning isolated AI-generated modules into synchronized, chat-accessible instruments.
-
We Need Strong Preconditions For Using Simulations In Policy
Societal-scale LLM agent simulations for policy need three preconditions: avoid neutral treatment of marginalized population simulations, require population participation, ensure accountability, plus development and deployment reports.
- HeadRoom: Lightweight, Edge-deployable Pipeline for Adaptive Notification Routing