Line of inquiry
Inquiring lines›How does AI reshape human reasonin…›How does AI reshape human skill, a…›this line of inquiry
How can AI agents autonomously learn and transfer skills across tasks?
A broader line of inquiry — a family of 51 specific questions the research asks around this. Follow one into its inquiring-line page, or move sideways to a related line below.
Questions in this line of inquiry 51
Specific inquiring lines the field asks around this — ordered from the most general framing down to the most specific angle.
- Can individual skills improve through reuse and accumulate experience across tasks?
- Can agentic reasoning outperform rigid rule-based systems for skill refinement?
- Can agent-authored skill libraries compound autonomy gains over time?
- Do learned workflows transfer between different agents with minimal accuracy loss?
- Can skill repositories evolve toward execution-oriented refinement over time?
- Can agent skills move from prompts to trainable parameters?
- Why do AI agents struggle with novel experiments but excel at routine tasks?
- Can agents teach each other skills without human supervision?
- Can agents improve from deployment signals without explicit human annotation?
- How does real tool integration change what agents learn compared to simulated tools?
- How can agents evolve their own skills without human input?
- How do agents automatically generate suitable learning tasks based on current capability?
- What infrastructure decouples generation from training in asynchronous agent loops?
- Why do self-improving agents concentrate progress in the fast non-parametric loop?
- Do weight-space skills lose detail compared to textual skill descriptions?
- What role does self-learning play in improving agent reasoning without annotation?
- Can self-improving agents become truly autonomous without intrinsic metacognition?
- Can process supervision improve agentic RL through meta-reasoning rewards?
- Which AI interaction patterns preserve learning while which ones degrade skill formation?
- What training method supports dynamic tool discovery in long-horizon agents?
- Does self-play feedback improve skills created from the agent's own experience?
- Can small numbers of curated demonstrations produce emergent agentic behavior?
- What makes trajectory quality matter more than one-shot task success?
- Can RL-trained meta-agents match or exceed manually designed workflows?
- Can agents learn to use scaffolding structure the way they learn token weights?
- Can a progressively stricter evaluator act like a curriculum for improving agents?
- Can curriculum approaches teach agents when to stop exploring?
- Should we train the evolver or the executor when building self-improving agents?
- Can next-state supervision work across different agent interaction types like conversations and tool calls?
- What stops evolved agent behaviors from generalizing beyond specific tasks?
- Can applicability conditions be preserved automatically when agents reflect on trials?
- How can agents learn when silence is better than intervention?
- What domain properties determine whether causal rules transfer to new agents?
- How do task stream groupings provide long-horizon learning signals for curation decisions?
- Can combinational creativity alone drive open-ended learning in agents?
- Can tool adaptation work without freezing the agent in the loop?
- Does the 78-demonstration principle apply to other AI capabilities beyond agency?
- What happens when agents interact with environments and learn from their own mistakes?
- Can simulation fidelity limit what agents learn from trained world models?
- Can agentic AI tools deliver productivity gains on learning tasks differently?
- Does bounding textual edits prevent skill degradation better than free rewriting?
- Can skill libraries prevent redundant narrow artifacts from proliferating?
- How should AI skills be created and managed like software artifacts?
- Why do current metacognitive training loops fail when agents encounter new domains?
- Can graph topology represent successful trajectory clusters more effectively than skill libraries?
- How do fast and slow timescales enable continual agent adaptation?
- What capabilities can emerge from self-modification that the original agent lacked?
- Why does delegation training help models that work alone?
- What specific qualities make some demonstrations more effective for agency training?
- How do agent capabilities change across 25 relay rounds of interaction?
- Can curator modules trained on one executor transfer to entirely different agent backbones?