TS
Tobias Schnabel
cs.LGcs.IRcs.AIcs.CLcs.HCcs.FLcs.SIecon.EMstat.ML
On Valency
published · living versionsW_mxude697·v1 · currentpublished
Recommendations as Treatments: Debiasing Learning and Evaluation
with Adith Swaminathan, Ashudeep Singh, Navin Chandak, Thorsten Joachims
1 version
Preprints & journals
22 papers in the corpus · 2015–2026LLMs Corrupt Your Documents When You Delegate2604.15597v1 · Philippe Laban, Tobias Schnabel, Jennifer Neville2026 · 1 citationarXiv
Reasoning about Reasoning: BAPO Bounds on Chain-of-Thought Token Complexity in LLMs2602.02909v2 · Kiran Tomlinson, Tobias Schnabel, Adith Swaminathan et al.2026 · 0 citationsarXiv
A Course Correction in Steerability Evaluation: Revealing Miscalibration and Side Effects in LLMs2505.23816v2 · Trenton Chang, Tobias Schnabel, Adith Swaminathan et al.2025 · 1 citationarXiv
Lost in Transmission: When and Why LLMs Fail to Reason Globally2505.08140v5 · Tobias Schnabel, Kiran Tomlinson, Adith Swaminathan et al.2025 · 0 citationsarXiv
On Overcoming Miscalibrated Conversational Priors in LLM-based Chatbots2406.01633v1 · Christine Herlihy, Jennifer Neville, Tobias Schnabel et al.2024 · 1 citationarXiv
Symbolic Prompt Program Search: A Structure-Aware Approach to Efficient Compile-Time Prompt Optimization2404.02319v2 · Tobias Schnabel, Jennifer Neville2024 · 7 citationsarXiv
When Newer is Not Better: Does Deep Learning Really Benefit Recommendation From Implicit Feedback?2305.01801v1 · Yushun Dong, Jundong Li, Tobias Schnabel2023 · 15 citationsarXiv
Situating Recommender Systems in Practice: Towards Inductive Learning and Incremental Updates2211.06365v1 · Tobias Schnabel, Mengting Wan, Longqi Yang2022 · 1 citationarXiv
Where Do We Go From Here? Guidelines For Offline Recommender Evaluation2211.01261v1 · Tobias Schnabel2022 · 1 citationarXiv
EvalRS: a Rounded Evaluation of Recommender Systems2207.05772v2 · Jacopo Tagliabue, Federico Bianchi, Tobias Schnabel et al.2022 · 0 citationsarXiv
Lightweight Compositional Embeddings for Incremental Streaming Recommendation2202.02427v1 · Mengyue Hang, Tobias Schnabel, Longqi Yang et al.2022 · 3 citationsarXiv
SummaC: Re-Visiting NLI-based Models for Inconsistency Detection in Summarization2111.09525v1 · Philippe Laban, Tobias Schnabel, Paul N. Bennett et al.2021 · 275 citationsarXiv
Keep it Simple: Unsupervised Simplification of Multi-Paragraph Text2107.03444v1 · Philippe Laban, Tobias Schnabel, Paul Bennett et al.2021 · 30 citationsAssociation for Computational Linguistics (2021)
Deep Generalized Method of Moments for Instrumental Variable Analysis1905.12495v2 · Andrew Bennett, Nathan Kallus, Tobias Schnabel2019 · 59 citationsAdvances in Neural Information Processing Systems 32 (2019) 3564--3574
Improving Recommender Systems Beyond the Algorithm1802.07578v1 · Tobias Schnabel, Paul N. Bennett, Thorsten Joachims2018 · 14 citationsarXiv
A preference elicitation interface for collecting dense recommender datasets with rich user information1706.08184v2 · Pantelis P. Analytis, Tobias Schnabel, Stefan Herzog et al.2017 · 2 citationsarXiv
Unbiased Learning-to-Rank with Biased Feedback1608.04468v1 · Thorsten Joachims, Adith Swaminathan, Tobias Schnabel2016 · 528 citationsarXiv
Unbiased Comparative Evaluation of Ranking Functions1604.07209v1 · Tobias Schnabel, Adith Swaminathan, Peter Frazier et al.2016 · 25 citationsarXiv
Online Updating of Word Representations for Part-of-Speech Tagging1604.00502v1 · Wenpeng Yin, Tobias Schnabel, Hinrich Schutze2016 · 10 citationsarXiv
Recommendations as Treatments: Debiasing Learning and Evaluation1602.05352v2 · Tobias Schnabel, Adith Swaminathan, Ashudeep Singh et al.2016 · 268 citationsarXivon Valency
Towards a Better Understanding of Predict and Count Models1511.02024v1 · S. Sathiya Keerthi, Tobias Schnabel, Rajiv Khanna2015 · 4 citationsarXiv
Using Shortlists to Support Decision Making and Improve Recommender System Performance1510.07545v2 · Tobias Schnabel, Paul N. Bennett, Susan T. Dumais et al.2015 · 34 citationsarXiv
Career total: 48 works. 22 are in this corpus.
Profile built from the corpus for this byline.
Author records are still filling in while the Hub is in alpha. If this is your page, you'll be able to claim it soon. Spot a mistake? Tell us.