Task success alone does not specify which behaviors post-training should reinforce. Even successful teacher trajectories can contain unsupported assumptions, unproductive repetition, or superficial verification. We propose Skill-FT (Skill-Guided Fine-Tuning), which uses natural-language skills to suppress undesirable habits and elicit underrepresented effective behaviors during teacher rollouts, then fine-tunes a target model on the resulting trajectories.
Is this you? As a journalist, you can create a free Muck Rack account to customize your profile, list your contact preferences, and upload a portfolio of your best work.
Claim your profile
Articles by Wayne Zhao
Computer Science > Computation and Language
Recently, linear attention layers have been increasingly adopted to replace softmax attention at scale for long-context modeling. However, existing context extension approaches typically apply continued pretraining directly without modifying these layers, overlooking the spectral properties of linear attention state dynamics.
Computer Science > Computation and Language
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website. Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them. Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
[2504.06792] Domain-Specific Pruning of Large Mixture-of-Experts Models with Few-shot Demonstrations
Work on one of the world's most important websites and make an impact on open science.
Deriving Vocal Fold Oscillation Information from Recorded Voice Signals Using Models of Phonation
This is an early access version, the complete PDF, HTML, and XML versions will be available soon. Open AccessArticle by Wayne Zhao 1 and Rita Singh 2,* 1 Department of Electrical and Computer Engineering, Carnegie Mellon University, Pittsburgh, PA 15213, USA 2 School of Computer Science, Carnegie Mellon University, Pittsburgh, PA 15213, USA * Author to whom correspondence should be addressed.
RocketQAv2: A Joint Training Method for Dense Passage Retrieval and Passage Re-ranking
[Submitted on 14 Oct 2021 ( v1 ), last revised 23 Apr 2023 (this version, v2)] Title:RocketQAv2: A Joint Training Method for Dense Passage Retrieval and Passage Re-ranking Download a PDF of the paper titled RocketQAv2: A Joint Training Method for Dense Passage Retrieval and Passage Re-ranking, by Ruiyang Ren and 6 other authors Download PDF Abstract: In various natural language processing tasks, passage retrieval and passage re-ranking are two key procedures in finding and ranking relevant...
Two Sessions signal Chinese government support for blockchain
Mar 21, 2019 |In With Chinese Characteristics, Views 2 min read (Image credit: Bigstock/LuckyStep48) Blockchain was on the agenda at the annual Two Sessions meeting of China’s legislature. Delegates and political leaders also discussed cryptocurrency, calling for efficient, clear and strict regulations. For non-currency blockchain, representatives expressed enthusiasm about a wide range of applications, which will likely receive state support. We count nine applications specifically named.
Show More
loading
Actions
Is this you?
As a journalist, you can create a free Muck Rack account to customize your profile, list your contact preferences, and upload a portfolio of your best work.Get in touch with Wayne
Contact Wayne, search articles and posts on X, monitor coverage, and track replies from one place.
Learn more about Muck Rack