PaperHub
影响力指数
98.01/100
前 0.1%
全站排名 #51
发表论文53 篇
平均评分5.4
年均产出17.7 篇/年

Bryan Kian Hsiang Low

Associate Professor@National University of Singapore·新加坡·OpenReview
研究方向

Data-centric AI · data valuation · data selection · Incentives in federated learning · Bayesian optimization · Gaussian process · active learning

6.7
11

INO-SGD: Addressing Utility Imbalance under Individualized Differential Privacy

ICLR 2026Poster
通讯
6.5
19

MEM1: Learning to Synergize Memory and Reasoning for Efficient Long-Horizon Agents

ICLR 2026Poster
6.0
17

DUET: Optimizing LLM Training Data Mixtures via Noisy Feedback from Unseen, Downstream Evaluation Tasks

ICLR 2026Poster
通讯
6.0
17

How to Cure Newton for Unlearning Neural Networks? An Empirical Study from the Hessian Perspective

ICLR 2026Poster
通讯
5.5
15

TreeGrad-Ranker: Feature Ranking via $O(L)$-Time Gradients for Decision Trees

ICLR 2026Poster
三作
5.5
14

ActiveDPO: Active Direct Preference Optimization for Sample-Efficient Alignment

ICLR 2026Poster
通讯
5.3
21

WaterDrum: Watermark-based Data-centric Unlearning Metric

ICLR 2026Poster
通讯
4.7
16

Configuring Parallel Training of Neural Networks using Bayesian Optimization

ICLR 2026Rejected
通讯
4.5
15

Diffusion-Inspired Reconfiguration of Transformers for Uncertainty Quantification

ICLR 2026Rejected
4.5
18

Efficient Stochastic Algorithms for Continual Finite-Sum Minimax Optimization

ICLR 2026Rejected
通讯
4.5
25

The Chicken and Egg Dilemma: Co-optimizing Data and Model Configurations for LLMs

ICLR 2026Rejected
通讯
4.0
16

REFRAG: Rethinking RAG based Decoding

ICLR 2026Rejected
三作
4.0
24

Prompts to Proxies: Emulating Human Preferences via a Compact LLM Ensemble

ICLR 2026Rejected
通讯
3.5
10

Adaptive Test-Time Compute Allocation via Training-Free Difficulty Proxies

ICLR 2026Rejected
通讯
3.5
13

COBRA: Contextual Bandit Algorithm for Ensuring Truthful Strategic Agents

ICLR 2026Rejected
通讯
3.5
6

Provably Sample-Efficient Active Preference Data Collection

ICLR 2026Rejected
通讯
-1

Uncertainty Quantification for Multimodal Large Language Models with Coherence-adjusted Semantic Volume

ICLR 2026Withdrawn
通讯
7.3
23

Broaden your SCOPE! Efficient Multi-turn Conversation Planning for LLMs with Semantic Space

ICLR 2025Spotlight
通讯
7.2
15

NICE Data Selection for Instruction Tuning in LLMs with Non-differentiable Evaluation Metric

ICML 2025Poster
通讯
7.0
21

PIED: Physics-Informed Experimental Design for Inverse Problems

ICLR 2025Poster
通讯
7.0
19

ReasonIR: Training Retrievers for Reasoning Tasks

COLM 2025Poster
6.8
20

Efficient Top-m Data Values Identification for Data Selection

ICLR 2025Poster
通讯
6.4
21

Incentivizing Time-Aware Fairness in Data Sharing

NeurIPS 2025Poster
通讯
6.3
11

Ferret: Federated Full-Parameter Tuning at Scale for Large Language Models

ICML 2025Poster
6.3
23

Neural Dueling Bandits: Preference-Based Optimization with Human Feedback

ICLR 2025Poster
通讯
6.0
25

Group-robust Sample Reweighting for Subpopulation Shifts via Influence Functions

ICLR 2025Poster
通讯
5.5
11

BILBO: BILevel Bayesian Optimization

ICML 2025Poster
三作
5.4
26

BILBO: BILevel Bayesian Optimization

ICLR 2025Rejected
三作
5.4
34

Source Attribution for Large Language Model-Generated Data

ICLR 2025Rejected
通讯
5.3
25

Data Value Estimation on Private Gradients

ICLR 2025Rejected
通讯
4.7
4

Ferret: Federated Full-Parameter Tuning at Scale for Large Language Models

ICLR 2025Withdrawn
4.6
19

Prompt Optimization with Human Feedback

ICLR 2025Rejected
通讯
3.7
16

Achieving Exact Federated Unlearning with Improved Post-Unlearning Performance

ICLR 2025Rejected
通讯
3.0
6

Global-to-Local Support Spectrums for Model Explainability

ICLR 2025Rejected
三作
合作者 (20)