Hard Negative Mining

Follow this topic Ask about this topic

Momentum

1 paper in the last four weeks, with none the four weeks before. 0.0% of all new papers.

Jul 13Week of Sep 28

Latest papers 18

All topics

Hard-Region Supervision: #1 on the Waymo Open Dataset 2D Video Panoptic Segmentation Leaderboard
Sep 30, 2026Jinghan YangPanoptic Segmentation Video Object Segmentation
HNR-DAC: Hard-Negative Reranking and Distribution-Aligned Classification for Scientific Claim Verification
Aug 7, 2026Zhenchao Wang, Xin Chen, Luoxi Zhang +2Claim Verification Distribution Shift
Southern University of Science and Technology, Shenzhen, China · Institute of Artificial Intelligence, Shenzhen University of Advanced Technology, Shenzhen, China · Shenzhen Institutes of Advanced Technology, Chinese Academy of Sciences, Shenzhen, China
Semantic Hardness Is Not Visual Hardness: Sign-Aware Hard Negative Mining for Sign Language Retrieval
Jul 10, 2026Junmyeong Lee, Chan Hur, ChangSu Choi +5Hard Negative Mining Cross-Modal Retrieval
School of Computing1 · ETRI Medical Informatics Laboratory4 · Graduate School of Culture Technology2 +1
Real-Time Hard Negative Sampling via LLM-based Clustering for Large-Scale Two-Tower Retrieval
Jul 1, 2026Ivan Ji, Liuyi Hu, Harrison +6Hard Negative Mining Dense Retrieval
Zihao · 1Meta · Xiangjun
Cross-Platform Chinese Offensive Comment Detection via Dual-Threshold Hard Example Mining
Jun 26, 2026Ruixing Ren, Junhui Zhao, Fangfang WangFew-Shot Domain Adaptation Domain Adaptation
School of Electronic and Information Engineering, Beijing Jiaotong University, Beijing 100044, China
Aligning Implied Statements for Implicit Hate Speech Generalizability with Context-Bounded Semi-hard Negative Mining
Jun 17, 2026Wicaksono Leksono Muhamad, Yunita SariContrastive Learning Hard Negative Mining
Mantera Studio · Universitas Gadjah Mada
Synthetic Hallucinations, Real Gains: Hard Negatives from Frontier Models for FIM Hallucination Mitigation
Jun 2, 2026Mahdi Erfanian, Nelson Daniel Troncoso, Aashna Garg +4LLM Hallucination Mitigation Hard Negative Mining
University of Illinois Chicago · Microsoft
When Hard Negatives Hurt: Bridging the Generative-Discriminative Gap in Hard Negative Synthesis for Retrieval
May 31, 2026Zhicheng Zhang, Jiwei Tang, Kuicai Dong +9Contrastive Learning Hard Negative Mining
Shenzhen International Graduate School, Tsinghua University Shenzhen, China · Huawei Technologies Co., Ltd. Shenzhen, China · City University of Hong Kong Hong Kong SAR, China +4
HELEA: Hard-Negative Benchmark and LLM-based Reranking for Robust Entity Alignment
May 27, 2026Yoonjin Jang, Junwoo Kim, Youngjoong KoHard Negative Mining
SungKyunKwan University Republic of Korea
Debiased Negative Mining Improves Out-of-distribution Detection with Pre-trained Vision-Language Models
May 22, 2026Bo Peng, Jie Lu, Guangquan Zhang +1Hard Negative Mining OOD Detection
University of Technology Sydney
SAGE: Scalable Automatic Gating Ensemble for Confident Negative Harvesting in Fraud Detection
May 19, 2026Sudheer Tubati, Amit GoyalEnsemble Learning Hard Negative Mining
Amazon Music Seattle, USA · Amazon Music San Francisco, USA
HNC: Leveraging Hard Negative Captions towards Models with Fine-Grained Visual-Linguistic Comprehension Capabilities
May 6, 2026Esra Dönmez, Pascal Tilli, Hsiu-Yu Yang +2Cross-Modal Learning Image Matching
Institute for Natural Language Processing, University of Stuttgart
Self-Mined Hardness for Safety Fine-Tuning
May 4, 2026Prakhar Gupta, Garv Shah, Donghua ZhangHard Negative Mining LLM Fine-Tuning
University of Michigan
DynamicPO: Dynamic Preference Optimization for Recommendation
May 1, 2026Xingyu Hu, Kai Zhang, Jiancan Wu +7Hard Negative Mining Preference Optimization
University of Science and Technology of China, Hefei, China · Shanghai Innovation Institute, Shanghai, China · Meituan, Chengdu, China
Selective Contrastive Learning For Gloss Free Sign Language Translation
Apr 24, 2026Changhao Lai, Rui Zhao, Xuewen Zhong +2Contrastive Learning Sign Language Translation
School of Informatics, Xiamen University, China · 2Key Lab of Digital Protection and Intelligent Processing of Intangible Cultural Heritage of Fujian-Taiwan (XMU), Ministry of Culture and Tourism, China · 3National Language Resources Monitoring and Research Center for Education and Teaching Media, Xiamen University, China
Co-Evolving Agents: Learning from Failures as Hard Negatives
Nov 27, 2025Yeonsung Jung, Trilok Padhi, Sina Shaham +4Hard Negative Mining Self-Improving Agents
KAIST · Georgia State University · Independent Researcher +2
SynCo: Synthetic Hard Negatives for Contrastive Visual Representation Learning
Oct 3, 2024Nikos Giakoumoglou, Tania StathakiContrastive Learning Hard Negative Mining
Imperial College London London, UK, SW7 2AZ
SalamandraTA at WMT 2026 Terminology Shared Task: Hard Examples Are Better Teachers
Date pendingXixian Liao, Maite MeleroHard Negative Mining LLM Fine-Tuning
Barcelona Supercomputer Center

Sep 30, 2026, cs.CVJ/K move · Enter open · S save

Hard-Region Supervision: #1 on the Waymo Open Dataset 2D Video Panoptic Segmentation Leaderboard

Jinghan Yang

Panoptic Segmentation Video Object Segmentation Hard Negative Mining Autonomous Driving

Open paper AskPDF arXiv

We describe our winning entry to the Waymo Open Dataset 2D Video Panoptic Segmentation Challenge. The task asks for a semantic class at every pixel of every frame and, for countable objects, an identity that holds across 100 frames and across five overlapping cameras. We build on DVIS++, a cascade of a segmenter, a tracker, and a refiner, as our baseline. We propose hard region supervision (HRS) to improve the baseline. In particular, we use the baseline to define the hard region as where it makes mistakes, and design a loss and an auxiliary prediction head for this region. The auxiliary head is used only in training and removed at test time, so at inference the model trained with HRS has the same architecture as the baseline. In addition, we propose three test-time steps that further improve the results: a two-model ensemble, a merge of the segmenter's output into the final panoptic map, and cross-camera identity linking. On the challenge test set, our entry reaches 0.3547 wSTQ, 0.2071 wAQ, and 0.6075 mIoU, ranking first on all three metrics. It is 3.6 wSTQ points ahead of the second entry and 2.4 points ahead of our DVIS++ baseline.

Momentum

Related topics

Latest papers 18