TC
Tong Che
cs.LGcs.AIcs.CVcs.CLcs.ROstat.MLcs.DCcs.NEcs.MMcs.SE
On Valency
published · living versionsW_bh98kykm·v1 · currentpublished
Short-to-Long: Detached Features for Long-Horizon Value Learning
with Boris Ivanovic, Marco Pavone
1 version
Preprints & journals
40 papers in the corpus · 2015–2026Short-to-Long: Detached Features for Long-Horizon Value Learning2610.30461v1 · Tong Che, Boris Ivanovic, Marco Pavone2026 · 0 citationsarXivon Valency
Shared Gaussianization: What Gaussian Regularizers Certify About Contrastive Learning, and What They Miss2610.10299v1 · Ruoyu Zhao, Yuting Chen, Jinheng Zhang et al.2026 · 0 citationsarXiv
Does the Model Use the Feature? Separating Steering from Mechanism in LLMs2610.07270v1 · Tong Che, Yilong Li2026 · 0 citationsarXiv
Second-Order Muon Done Right: A Principled Marriage of Spectral Geometry and Curvature2608.09763v1 · Tong Che2026 · 0 citationsarXiv
Wait-Free Locks Should Not Fear Later Arrivals2607.16571v1 · Tong Che2026 · 0 citationsarXiv
Learning with a Single Rollout via Pass@k Policy Optimization2606.25451v2 · Fengdi Che, Yang Liu, Lei Yu et al.2026 · 0 citationsarXiv
Greed Is Learned: Visible Incentives as Reward-Hacking Triggers2606.16914v2 · Tong Che, Rui Wu2026 · 0 citationsarXiv
Constitutional Value Potentials: reading and steering internal priority margins in language models2606.15420v2 · Tong Che, Rui Wu2026 · 0 citationsarXiv
EMBER: Efficient Memory via Budgeted Evidence Retention for Long-Horizon Agents2606.05894v2 · Yilong Li, Suman Banerjee, Tong Che2026 · 0 citationsarXiv
Reference Feature Atlases for Mechanistic Auditing of Language Models2607.22570v1 · Rui Wu, Tong Che2026 · 0 citationsarXiv
Cast a Wider Net: Coordinated Pass@K Policy Optimization for Code Reasoning2605.27000v3 · Yilong Li, Suman Banerjee, Tong Che2026 · 0 citationsarXiv
Reasoning over Precedents Alongside Statutes: Case-Augmented Deliberative Alignment for LLM Safety2601.08000v1 · Can Jin, Rui Wu, Tong Che et al.2026 · 0 citationsarXiv
Alpamayo-R1: Bridging Reasoning and Action Prediction for Generalizable Autonomous Driving in the Long Tail2511.00088v2 · NVIDIA: Yan Wang, Wenjie Luo, Junjie Bai et al.2025 · 1 citationarXiv
AIRL-S: Unifying Reinforcement Learning and Search-Based Test-Time Scaling via Adversarial Inverse Reinforcement Learning2508.14313v5 · Can Jin, Yang Zhou, Qixin Zhang et al.2025 · 0 citationsarXiv
Two Heads are Better Than One: Test-time Scaling of Multi-agent Collaborative Reasoning2504.09772v4 · Can Jin, Hongwu Peng, Qixin Zhang et al.2025 · 0 citationsarXiv
LoR-VP: Low-Rank Visual Prompting for Efficient Vision Model Adaptation2502.00896v3 · Can Jin, Ying Li, Mingyu Zhao et al.2025 · 0 citationsarXiv
Learning Multiple Initial Solutions to Optimization Problems2411.02158v2 · Elad Sharony, Heng Yang, Tong Che et al.2024 · 0 citationsarXiv
LLaMA-Berry: Pairwise Optimization for O1-like Olympiad-Level Mathematical Reasoning2410.02884v2 · Di Zhang, Jianbo Wu, Jingdi Lei et al.2024 · 3 citationsarXiv
Parallelized Spatiotemporal Binding2402.17077v1 · Gautam Singh, Yue Wang, Jiawei Yang et al.2024 · 0 citationsarXiv
Learning from Teaching Regularization: Generalizable Correlations Should be Easy to Imitate2402.02769v3 · Can Jin, Tong Che, Hongwu Peng et al.2024 · 17 citationsarXiv
EmerNeRF: Emergent Spatial-Temporal Scene Decomposition via Self-Supervision2311.02077v1 · Jiawei Yang, Boris Ivanovic, Or Litany et al.2023 · 27 citationsarXiv
Bayesian Reparameterization of Reward-Conditioned Reinforcement Learning with Energy-based Models2305.11340v1 · Wenhao Ding, Tong Che, Ding Zhao et al.2023 · 0 citationsarXiv
SPE: Symmetrical Prompt Enhancement for Fact Probing2211.07078v1 · Yiyuan Li, Tong Che, Yezhen Wang et al.2022 · 5 citationsarXiv
Foundation Models for Semantic Novelty in Reinforcement Learning2211.04878v1 · Tarun Gupta, Peter Karkus, Tong Che et al.2022 · 1 citationarXiv
Guided Conditional Diffusion for Controllable Traffic Simulation2210.17366v1 · Ziyuan Zhong, Davis Rempe, Danfei Xu et al.2022 · 126 citationsarXiv
Robust and Controllable Object-Centric Learning through Energy-based Models2210.05519v1 · Ruixiang Zhang, Tong Che, Boris Ivanovic et al.2022 · 0 citationsarXiv
Your Autoregressive Generative Model Can be Better If You Treat It as an Energy-Based One2206.12840v1 · Yezhen Wang, Tong Che, Bo Li et al.2022 · 0 citationsarXiv
Sparse Mixture-of-Experts are Domain Generalizable Learners2206.04046v6 · Bo Li, Yifei Shen, Jingkang Yang et al.2022 · 23 citationsarXiv
Energy-Based Open-World Uncertainty Modeling for Confidence Calibration2107.12628v3 · Yezhen Wang, Bo Li, Tong Che et al.2021 · 54 citationsarXiv
AUTO3D: Novel view synthesis through unsupervisely learned variational viewpoint and global 3D representation2007.06620v2 · Xiaofeng Liu, Tong Che, Yiqun Lu et al.2020 · 23 citationsarXiv
Rethinking Distributional Matching Based Domain Adaptation2006.13352v2 · Bo Li, Yezhen Wang, Tong Che et al.2020 · 38 citationsarXiv
Your GAN is Secretly an Energy-based Model and You Should use Discriminator Driven Latent Sampling2003.06060v3 · Tong Che, Ruixiang Zhang, Jascha Sohl-Dickstein et al.2020 · 49 citationsarXiv
Deep Verifier Networks: Verification of Deep Discriminative Models with Deep Generative Models1911.07421v3 · Tong Che, Xiaofeng Liu, Site Li et al.2019 · 53 citationsarXiv
Conservative Wasserstein Training for Pose Estimation1911.00962v1 · Xiaofeng Liu, Yang Zou, Tong Che et al.2019 · 37 citationsarXiv
Residual Connections Encourage Iterative Inference1710.04773v2 · Stanis\law Jastrzebski, Devansh Arpit, Nicolas Ballas et al.2017 · 77 citationsarXiv
Boundary-Seeking Generative Adversarial Networks1702.08431v4 · R Devon Hjelm, Athul Paul Jacob, Tong Che et al.2017 · 144 citationsarXiv
Maximum-Likelihood Augmented Discrete Generative Adversarial Networks1702.07983v1 · Tong Che, Yanran Li, Ruixiang Zhang et al.2017 · 168 citationsarXiv
Mode Regularized Generative Adversarial Networks1612.02136v5 · Tong Che, Yanran Li, Athul Paul Jacob et al.2016 · 442 citationsarXiv
Architectural Complexity Measures of Recurrent Neural Networks1602.08210v3 · Saizheng Zhang, Yuhuai Wu, Tong Che et al.2016 · 131 citationsarXiv
Stabilizing Logs for Eventually Linearizable Shared Objects1512.08258v3 · Tong Che2015 · 0 citationsarXiv
Career total: 53 works. 40 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.