Learning desk

MultiAgent EDU StackGather good sources. Teach what matters.
T3Mermaid to ASCII art (mermaid-ascii)T3Kimi K3, and what we can still learn from the pelican benchmarkT3Firefox in WebAssemblyT3Spot birds not golfT3[AINews] Kimi K3 2.8T-A50B: the largest open model ever released; Opus 4.8-class at Sonnet 5 pricingT1From physical surfaces to human-centric heat stress: LST and UTCI heat mapping reveals nonlinear effects of urban morphologyT1DualHNIE: Dual-Channel Hypergraph Learning for Node Importance Estimation in Heterogeneous Knowledge GraphsT1GenTL: A General Transfer Learning Model for Building Thermal DynamicsT1A short review on the maximum clique problem algorithms with classical, AI, and quantum methodsT1Man, Machine, and Masterpiece: Artistic Ownership in the AI EraT1HABIB_TAZ at SemEval-2026 Task 11: Disentangling Formal Logic from Content via Synthetic Training and Multi-Objective OptimizationT1How Well Does AI-Generated Feedback Work? Intrinsic and Extrinsic Evaluation across more than 20,000 EFL Essay DraftsT3Mermaid to ASCII art (mermaid-ascii)T3Kimi K3, and what we can still learn from the pelican benchmarkT3Firefox in WebAssemblyT3Spot birds not golfT3[AINews] Kimi K3 2.8T-A50B: the largest open model ever released; Opus 4.8-class at Sonnet 5 pricingT1From physical surfaces to human-centric heat stress: LST and UTCI heat mapping reveals nonlinear effects of urban morphologyT1DualHNIE: Dual-Channel Hypergraph Learning for Node Importance Estimation in Heterogeneous Knowledge GraphsT1GenTL: A General Transfer Learning Model for Building Thermal DynamicsT1A short review on the maximum clique problem algorithms with classical, AI, and quantum methodsT1Man, Machine, and Masterpiece: Artistic Ownership in the AI EraT1HABIB_TAZ at SemEval-2026 Task 11: Disentangling Formal Logic from Content via Synthetic Training and Multi-Objective OptimizationT1How Well Does AI-Generated Feedback Work? Intrinsic and Extrinsic Evaluation across more than 20,000 EFL Essay Drafts
← Dispatches

Predictive Human Preference: From Model Ranking to Model Routing

Named practitioner synthesis

#195

T3digested
Topic
Post-Training
First seen
2026-07-16 19:08:00
Last seen
2026-07-16 19:08:00

Source raw items (1)

  • Blog / Newsletter2026-07-16 19:07:19
    Predictive Human Preference: From Model Ranking to Model Routing

    <p>A challenge of building AI applications is choosing which model to use. What if we don’t have to? What if we can predict the best model for any prompt? Predictive human preference aims to predict which model users might prefer for a specific query.</p> <p>Human preference has emerged to be both the Northstar and a powerful tool for AI model development. Human preference guides post-training techniques including <a href="https://huyenchip.com/2023/05/02/rlhf.html">RLHF</a> and <a href="https://arxiv.org/abs/2305.18290">DPO</a>. Human preference is also used to rank AI models, as used by LMSYS’s <a href="https://arena.lmsys.org/">Chatbot Arena</a>.</p> <p>Chatbot Arena aims to determine which model is generally preferred. I wanted to see if it’s possible to predict which model is preferred <em>for each query</em>.</p> <p>One use case of predictive human preference is model routing. For example, if we know in advance that for a prompt, users will prefer Claude Instant’s response ove