EVE enables verifiable self-evolution of MLLMs by using a Challenger-Solver architecture to generate dynamic executable visual transformations that produce VQA problems with absolute execution-verified ground truth.
Self-Play Only Evolves When Self-Synthetic Pipeline Ensures Learnable Information Gain
6 Pith papers cite this work. Polarity classification is still indexing.
6
Pith papers citing it
abstract
Large language models (LLMs) make it plausible to build systems that improve through self-evolving loops, but many existing proposals are better understood as self-play and often plateau quickly. A central failure mode is that the loop synthesises more data without increasing learnable information for the next iteration. Through experiments on a self-play coding task, we reveal that sustainable self-evolution requires a self-synthesised data pipeline with learnable information that increases across iterations. We identify triadic roles that self-evolving LLMs play: the Proposer, which generates tasks; the Solver, which attempts solutions; and the Verifier, which provides training signals, and we identify three system designs that jointly target learnable information gain from this triadic roles perspective. Asymmetric co-evolution closes a weak-to-strong-to-weak loop across roles. Capacity growth expands parameter and inference-time budgets to match rising learnable information. Proactive information seeking introduces external context and new task sources that prevent saturation. Together, these modules provide a measurable, system-level path from brittle self-play dynamics to sustained self-evolution.
citation-role summary
method 1
citation-polarity summary
years
2026 6roles
method 1polarities
use method 1representative citing papers
OpenRCA 2.0 is the first cross-system RCA benchmark with step-wise causal annotations, revealing that 11 frontier LLMs achieve 20.7% exact root-cause recovery and struggle with causal grounding (61.5% vs 76.0% ungrounded).
Evolutionary coding agents achieve most benchmark gains through a small subset of edit types and by cycling previously deleted code lines rather than developing new algorithmic structures.
Experiments on coding and deterministic tasks demonstrate that data gating is sufficient for self-play stability while reward variants are not, revealing the Grounded Proposer Paradox and a two-stage phase transition under continuous gate strictness.
ErrorProbe introduces a self-improving pipeline for attributing semantic failures in LLM multi-agent systems to specific agents and steps via anomaly detection, backward tracing, and tool-grounded validation with verified episodic memory.
Randomly masking a proposer's output vocabulary during training and generation sustains curriculum diversity and improves solver accuracy by +4.4 points at 8B in LLM co-evolution.
citing papers explorer
-
EVE: Verifiable Self-Evolution of MLLMs via Executable Visual Transformations
EVE enables verifiable self-evolution of MLLMs by using a Challenger-Solver architecture to generate dynamic executable visual transformations that produce VQA problems with absolute execution-verified ground truth.
-
OpenRCA 2.0: From Outcome Labels to Causal Process Supervision
OpenRCA 2.0 is the first cross-system RCA benchmark with step-wise causal annotations, revealing that 11 frontier LLMs achieve 20.7% exact root-cause recovery and struggle with causal grounding (61.5% vs 76.0% ungrounded).
-
What Do Evolutionary Coding Agents Evolve?
Evolutionary coding agents achieve most benchmark gains through a small subset of edit types and by cycling previously deleted code lines rather than developing new algorithmic structures.
-
Survive or Collapse: The Asymmetric Roles of Data Gating and Reward Grounding in Self-Play RL
Experiments on coding and deterministic tasks demonstrate that data gating is sufficient for self-play stability while reward variants are not, revealing the Grounded Proposer Paradox and a two-stage phase transition under continuous gate strictness.
-
Towards Self-Improving Error Diagnosis in Multi-Agent Systems
ErrorProbe introduces a self-improving pipeline for attributing semantic failures in LLM multi-agent systems to specific agents and steps via anomaly detection, backward tracing, and tool-grounded validation with verified episodic memory.
-
Vocabulary Dropout for Curriculum Diversity in LLM Co-Evolution
Randomly masking a proposer's output vocabulary during training and generation sustains curriculum diversity and improves solver accuracy by +4.4 points at 8B in LLM co-evolution.