SG
Surbhi Goel
cs.LGstat.MLcs.DScs.AIcs.CLcs.GTcs.CCmath.OCcs.CRcs.CV
On Valency
published · living versionsW_97xvna8t·v1 · currentpublished
Understanding and Mitigating Role Spoofing in LLMs
with Ezra Edelman, Benjamin Edelman, Eran Malach
1 version
Preprints & journals
52 papers in the corpus · 2016–2026Narrowing the Collaboration Gap, Probably2610.19215v1 · Mirah Shi, Marcel Hussing, Ira Globus-Harris et al.2026 · 0 citationsarXiv
LESSER: Post-Training Data Selection with Output-Layer Gradients2610.03702v1 · Lyuxin David Zhang, Eric Wong, Surbhi Goel et al.2026 · 0 citationsarXiv
MoRE: Scaling mixture of experts with hardware-aware low-rank routing2609.36301v1 · Honam Wong, Surbhi Goel, Enric Boix-Adsera2026 · 0 citationsarXiv
Delegating Authorization to Misaligned Agents: Coalitional Alignment and Safe Control2609.15803v1 · Natalie Collina, Surbhi Goel, Aaron Roth et al.2026 · 0 citationsarXiv
Is Code Better Than Language for Algorithmic Reasoning2606.15589v1 · Terry Tong, Yu Feng, Surbhi Goel et al.2026 · 0 citationsarXiv
Less Data, Faster Training: repeating smaller datasets speeds up learning via sampling biases2605.20314v2 · Jingwen Liu, Ezra Edelman, Surbhi Goel et al.2026 · 0 citationsarXiv
Learning When to Stop: Selective Imitation Learning Under Arbitrary Dynamics Shift2605.09183v2 · Surbhi Goel, Jonathan Pei, James Wang2026 · 0 citationsarXiv
Weight Clipping for Robust Conformal Inference under Unbounded Covariate Shifts2605.02072v1 · James Wang, Surbhi Goel2026 · 0 citationsarXiv
Model Agreement via Anchoring2602.23360v1 · Eric Eaton, Surbhi Goel, Marcel Hussing et al.2026 · 0 citationsarXiv
Reliable Abstention under Adversarial Injections: Tight Lower Bounds and New Upper Bounds2602.20111v1 · Ezra Edelman, Surbhi Goel2026 · 0 citationsarXiv
Personalization Aids Pluralistic Alignment Under Competition2602.13451v1 · Natalie Collina, Surbhi Goel, Aaron Roth et al.2026 · 0 citationsarXiv
Conformal Language Model Reasoning with Coherent Factuality2505.17126v1 · Maxon Rubin-Toles, Maya Gambhir, Keshav Ramji et al.2025 · 0 citationsarXiv
Probabilistic Stability Guarantees for Feature Attributions2504.13787v3 · Helen Jin, Anton Xue, Weiqiu You et al.2025 · 0 citationsarXiv
Collaborative Prediction: Tractable Information Aggregation via Agreement2504.06075v1 · Natalie Collina, Ira Globus-Harris, Surbhi Goel et al.2025 · 0 citationsarXiv
A Theory of Learning with Autoregressive Chain of Thought2503.07932v2 · Nirmit Joshi, Gal Vardi, Adam Block et al.2025 · 0 citationsarXiv
Testing Noise Assumptions of Learning Algorithms2501.09189v3 · Surbhi Goel, Adam R. Klivans, Konstantinos Stavropoulos et al.2025 · 0 citationsarXiv
Tractable Agreement Protocols2411.19791v1 · Natalie Collina, Surbhi Goel, Varun Gupta et al.2024 · 0 citationsarXiv
Logicbreaks: A Framework for Understanding Subversion of Rule-based Inference2407.00075v5 · Anton Xue, Avishree Khare, Rajeev Alur et al.2024 · 0 citationsarXiv
Explicitly Encoding Structural Symmetry is Key to Length Generalization in Arithmetic Tasks2406.01895v1 · Mahdi Sabbaghi, George Pappas, Hamed Hassani et al.2024 · 0 citationsarXiv
Tolerant Algorithms for Learning with Arbitrary Covariate Shift2406.02742v1 · Surbhi Goel, Abhishek Shetty, Konstantinos Stavropoulos et al.2024 · 1 citationarXiv
Stochastic Bandits with ReLU Neural Networks2405.07331v1 · Kan Xu, Hamsa Bastani, Surbhi Goel et al.2024 · 0 citationsarXiv
Complexity Matters: Dynamics of Feature Learning in the Presence of Spurious Correlations2403.03375v3 · GuanWen Qiu, Da Kuang, Surbhi Goel2024 · 0 citationsarXiv
The Evolution of Statistical Induction Heads: In-Context Learning Markov Chains2402.11004v1 · Benjamin L. Edelman, Ezra Edelman, Surbhi Goel et al.2024 · 8 citationsarXiv
Pareto Frontiers in Neural Feature Learning: Data, Compute, Width, and Luck2309.03800v2 · Benjamin L. Edelman, Surbhi Goel, Sham Kakade et al.2023 · 0 citationsarXiv
Learning Narrow One-Hidden-Layer ReLU Networks2304.10524v1 · Sitan Chen, Zehao Dou, Surbhi Goel et al.2023 · 1 citationarXiv
Transformers Learn Shortcuts to Automata2210.10749v2 · Bingbin Liu, Jordan T. Ash, Surbhi Goel et al.2022 · 13 citationsarXiv
Recurrent Convolutional Neural Networks Learn Succinct Learning Algorithms2209.00735v2 · Surbhi Goel, Sham Kakade, Adam Tauman Kalai et al.2022 · 1 citationarXiv
Hidden Progress in Deep Learning: SGD Learns Parities Near the Computational Limit2207.08799v3 · Boaz Barak, Benjamin L. Edelman, Surbhi Goel et al.2022 · 15 citationsarXiv
Understanding Contrastive Learning Requires Incorporating Inductive Biases2202.14037v1 · Nikunj Saunshi, Jordan Ash, Surbhi Goel et al.2022 · 24 citationsarXiv
Anti-Concentrated Confidence Bonuses for Scalable Exploration2110.11202v2 · Jordan T. Ash, Cyril Zhang, Surbhi Goel et al.2021 · 1 citationInternational Conference on Learning Representations 2022
Inductive Biases and Variable Creation in Self-Attention Mechanisms2110.10090v2 · Benjamin L. Edelman, Surbhi Goel, Sham Kakade et al.2021 · 19 citationsarXiv
Statistical Estimation from Dependent Data2107.09773v1 · Yuval Dagan, Constantinos Daskalakis, Nishanth Dikkala et al.2021 · 0 citationsarXiv
Investigating the Role of Negatives in Contrastive Representation Learning2106.09943v1 · Jordan T. Ash, Surbhi Goel, Akshay Krishnamurthy et al.2021 · 11 citationsarXiv
Gone Fishing: Neural Active Learning with Fisher Embeddings2106.09675v2 · Jordan T. Ash, Surbhi Goel, Akshay Krishnamurthy et al.2021 · 25 citationsarXiv
Acceleration via Fractal Learning Rate Schedules2103.01338v2 · Naman Agarwal, Surbhi Goel, Cyril Zhang2021 · 3 citationsarXiv
Tight Hardness Results for Training Depth-2 ReLU Networks2011.13550v1 · Surbhi Goel, Adam Klivans, Pasin Manurangsi et al.2020 · 5 citationsarXiv
From Boltzmann Machines to Neural Networks and Back Again2007.12815v1 · Surbhi Goel, Adam Klivans, Frederic Koehler2020 · 2 citationsarXiv
Statistical-Query Lower Bounds via Functional Gradients2006.15812v2 · Surbhi Goel, Aravind Gollakota, Adam Klivans2020 · 12 citationsarXiv
Superpolynomial Lower Bounds for Learning One-Layer Neural Networks using Gradient Descent2006.12011v2 · Surbhi Goel, Aravind Gollakota, Zhihan Jin et al.2020 · 19 citationsarXiv
Approximation Schemes for ReLU Regression2005.12844v2 · Ilias Diakonikolas, Surbhi Goel, Sushrut Karmalkar et al.2020 · 15 citationsarXiv
Efficiently Learning Adversarially Robust Halfspaces with Noise2005.07652v1 · Omar Montasser, Surbhi Goel, Ilias Diakonikolas et al.2020 · 9 citationsarXiv
Time/Accuracy Tradeoffs for Learning a ReLU with respect to Gaussian Marginals1911.01462v1 · Surbhi Goel, Sushrut Karmalkar, Adam Klivans2019 · 15 citationsarXiv
Learning Restricted Boltzmann Machines with Arbitrary External Fields1906.06595v1 · Surbhi Goel2019 · 0 citationsarXiv
Learning Mixtures of Graphs from Epidemic Cascades1906.06057v2 · Jessica Hoffmann, Soumya Basu, Surbhi Goel et al.2019 · 1 citationarXiv
Recovering the Lowest Layer of Deep Networks with High Threshold Activations1903.09231v2 · Surbhi Goel, Rina Panigrahy2019 · 0 citationsarXiv
Quantifying Perceptual Distortion of Adversarial Examples1902.08265v1 · Matt Jordan, Naren Manoj, Surbhi Goel et al.2019 · 30 citationsarXiv
Learning Ising Models with Independent Failures1902.04728v1 · Surbhi Goel, Daniel M. Kane, Adam R. Klivans2019 · 8 citationsarXiv
Improved Learning of One-hidden-layer Convolutional Neural Networks with Overlaps1805.07798v2 · Simon S. Du, Surbhi Goel2018 · 5 citationsarXiv
Learning One Convolutional Layer with Overlapping Patches1802.02547v1 · Surbhi Goel, Adam Klivans, Raghu Meka2018 · 41 citationsarXiv
Learning Neural Networks with Two Nonlinear Layers in Polynomial Time1709.06010v4 · Surbhi Goel, Adam Klivans2017 · 31 citationsarXiv
Career total: 60 works. 52 are in this corpus.Showing the 50 most recent.
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.