VisualCoding: Learning Executable Skills for Reliable Visual Reasoning
We propose VisualCoding, which combines execution-guided acquisition of reusable visual programs with direct adoption of their answers on applicable inputs.
ProofPaper ↗
Key points
- Learning reusable skills can improve visual reasoning agents, yet applying a skill still requires the agent to implement or invoke a procedure and produce the final answer.
- Execution-Guided Program Acquisition composes visual primitives and additional code into family-level solvers, then iteratively tests and revises them on labeled training instances.
- Scope-Aware Direct Execution runs a retained solver and returns its valid in-scope answer.
- We evaluate VisualCoding on four models across four benchmarks spanning 69 task families.
Sources (1)
- [1]VisualCoding: Learning Executable Skills for Reliable Visual ReasoningarXiv (AI, ML, NLP, CV, robotics, multi-agent) · Oct 9, 11:46 PM
We propose VisualCoding, which combines execution-guided acquisition of reusable visual programs with direct adoption of their answers on applicable inputs.
Learning reusable skills can improve visual reasoning agents, yet applying a skill still requires the agent to implement or invoke a procedure and produce the final answer.
Extractive summary: sentences quoted from the sources.
Before this
- Oct 9, 2026Does Location Matter? A Black-Box Causal Audit Reveals Model-Dependent Audio Grounding in LALM Hallucination
- Oct 9, 2026Work2Skill: Capturing Human GUI Experience for Reusable Agent Skill Induction
- Oct 8, 2026openai/codex rust-v0.162.0: 0.162.0
- Oct 8, 2026Microsoft Joins the Local AI Push
- Oct 8, 2026Understanding Jev, the new model everyone is talking about
- Oct 8, 2026DataVista: Diagnosing Multimodal LLMs on Data Video Understanding