Capability Conditioned Scaffolding for Professional Human LLM Collaboration
Summary
Introduces Capability Conditioned Scaffolding, a framework for LLM collaboration that adapts intervention based on user expertise domains to prevent Professional Domain Drift, with pilot evaluation on MMLU subsets.
View Cached Full Text
Cached at: 05/18/26, 06:31 AM
# Capability Conditioned Scaffolding for Professional Human LLM Collaboration Source: [https://arxiv.org/abs/2605.15404](https://arxiv.org/abs/2605.15404) [View PDF](https://arxiv.org/pdf/2605.15404) > Abstract:Large language model personalization typically adapts outputs to user preferences and style but does not account for differences in user evaluation capacity across domains of expertise\. This limitation can encourage Professional Domain Drift, where users rely on AI generated reasoning in domains they cannot reliably evaluate\. We introduce Capability Conditioned Scaffolding, a typed framework that partitions expertise into strong, mixed, and weak domains and conditions intervention behavior on structured capability profiles\. A pilot evaluation across multiple MMLU subsets and four LLM substrates shows consistent profile conditioned intervention behavior, including categorical inversion under profile swapping and selective activation in mixed domain risk zones\. These findings suggest that capability aware scaffolding can support more reliable professional human AI collaboration beyond stylistic personalization\. ## Submission history From: Sen Yang \[[view email](https://arxiv.org/show-email/414fc87a/2605.15404)\] **\[v1\]**Thu, 14 May 2026 20:42:03 UTC \(559 KB\)
Similar Articles
Capability Self-Assessment: Teaching LLMs to Know Their Limits
This paper introduces Capability Self-Assessment (CSA) for LLMs, formulating it as a policy-learning problem. Experiments show that reinforcement learning effectively teaches models to recognize their own limits and delegate queries they cannot solve, outperforming supervised fine-tuning and generalizing well out-of-distribution.
Scaffold-Mediated Post-Training: Co-Evolving Model Parameters and Procedural Scaffold Graphs
This paper proposes scaffold-mediated post-training, a paradigm where procedural scaffolds co-evolve with LLM parameters through discovery, distillation, and dynamic recompilation. On FeatureBench, automatically discovered skills improve pass rate by 8.1pp, with a 27.7% pass rate after distillation.
LLM Agents for Deliberative Collaboration: A Study on Joint Decision Making Under Partial Observability
This paper formalizes deliberative collaboration for LLM agents under partial observability, introduces a scalable benchmark across multiple domains, and systematically evaluates representative LLMs, finding that complex tasks remain challenging while deliberation can enable error correction.
From Memory to Skills: Evidence-Grounded Co-Evolution Governance for Long-Horizon LLM Agents
MSCE is a training-free framework that organizes LLM agent experience into three memory levels and converts them into reusable skills with evidence links, outperforming existing memory and skill-augmented baselines.
What happens if you design tools for LLMs instead of letting LLM use human tools ?
The blog post explores the rationale and benefits of designing specialized tools for LLMs instead of allowing them to use general-purpose human tools, emphasizing how LLMs' unique requirements and capability restrictions could enhance their power.