PaperHub
影响力指数
83.73/100
前 1.1%
全站排名 #676
发表论文28 篇
平均评分5.1
年均产出9.3 篇/年

Gagandeep Singh

Assistant Professor@University of Illinois, Urbana Champaign·OpenReview
研究方向

Adversarial Robustness · Neural Network Verification · Abstract Interpretation · Formal Methods

8.2
27

DINGO: Constrained Inference for Diffusion LLMs

NeurIPS 2025Poster
通讯
6.8
24

TRAP: Targeted Redirecting of Agentic Preferences

NeurIPS 2025Poster
三作
6.7
11

IterGen: Iterative Semantic-aware Structured LLM Generation with Backtracking

ICLR 2025Poster
6.3
8

CRANE: Reasoning with constrained LLM generation

ICML 2025Poster
通讯
6.3
21

Certifying Counterfactual Bias in LLMs

ICLR 2025Poster
通讯
6.0
32

UTF-8 Plumbing: Byte-level Tokenizers Unavoidably Enable LLMs to Generate Ill-formed UTF-8

COLM 2025Poster
三作
5.7
5

Support is All You Need for Certified VAE Training

ICLR 2025Poster
通讯
4.8
25

Stochastic Monkeys at Play: Random Augmentations Cheaply Break LLM Safety Alignment

ICLR 2025Withdrawn
通讯
4.8
15

ARQ: A Mixed-Precision Quantization Framework for Accurate and Certifiably Robust DNNs

ICLR 2025Rejected
4.5
22

Decoding Intelligence: A Framework for Certifying Knowledge Comprehension in LLMs

ICLR 2025Rejected
三作
4.5
16

Robust Thompson Sampling Algorithms Against Reward Poisoning Attacks

ICLR 2025Rejected
三作
4.2
7

Universal Black-Box Reward Poisoning Attack against Offline Reinforcement Learning

ICLR 2025Withdrawn
三作
3.8
5

Towards Universal Certified Robustness with Multi-Norm Training

ICLR 2025Withdrawn
二作
3.5
5

Binary Reward Labeling: Bridging Offline Preference and Reward-Based Reinforcement Learning

ICLR 2025Withdrawn
通讯
3.0
5

Two-Step Offline Preference-Based Reinforcement Learning with Constrained Actions

ICLR 2025Withdrawn
通讯
合作者 (20)