Search Results for author: Jiahao Xu

Found 20 papers, 13 papers with code

Thoughts Are All Over the Place: On the Underthinking of o1-Like LLMs

no code implementations30 Jan 2025 Yue Wang, Qiuzhi Liu, Jiahao Xu, Tian Liang, Xingyu Chen, Zhiwei He, Linfeng Song, Dian Yu, Juntao Li, Zhuosheng Zhang, Rui Wang, Zhaopeng Tu, Haitao Mi, Dong Yu

To address underthinking, we propose a decoding strategy with thought switching penalty TIP that discourages premature transitions between thoughts, encouraging deeper exploration of each reasoning path.

Do NOT Think That Much for 2+3=? On the Overthinking of o1-Like LLMs

no code implementations30 Dec 2024 Xingyu Chen, Jiahao Xu, Tian Liang, Zhiwei He, Jianhui Pang, Dian Yu, Linfeng Song, Qiuzhi Liu, Mengfei Zhou, Zhuosheng Zhang, Rui Wang, Zhaopeng Tu, Haitao Mi, Dong Yu

The remarkable performance of models like the OpenAI o1 can be attributed to their ability to emulate human-like long-time thinking during inference.

GSM8K

Teaching LLMs to Refine with Tools

no code implementations22 Dec 2024 Dian Yu, Yuheng Zhang, Jiahao Xu, Tian Liang, Linfeng Song, Zhaopeng Tu, Haitao Mi, Dong Yu

We propose CaP, a novel approach that uses external tools to refine chain-of-thought (CoT) responses generated by the same or other LLMs.

Findings of the WMT 2024 Shared Task on Discourse-Level Literary Translation

1 code implementation16 Dec 2024 Longyue Wang, Siyou Liu, Chenyang Lyu, Wenxiang Jiao, Xing Wang, Jiahao Xu, Zhaopeng Tu, Yan Gu, WeiYu Chen, Minghao Wu, Liting Zhou, Philipp Koehn, Andy Way, Yulin Yuan

Following last year, we have continued to host the WMT translation shared task this year, the second edition of the Discourse-Level Literary Translation.

Translation

Critical Tokens Matter: Token-Level Contrastive Estimation Enhances LLM's Reasoning Capability

1 code implementation29 Nov 2024 Zicheng Lin, Tian Liang, Jiahao Xu, Qiuzhi Lin, Xing Wang, Ruilin Luo, Chufan Shi, Siheng Li, Yujiu Yang, Zhaopeng Tu

Our results underscore the potential of leveraging critical tokens to reduce errors in reasoning tasks, advancing the development of AI systems capable of robust logical deduction.

GSM8K Math +1

Draft Model Knows When to Stop: A Self-Verification Length Policy for Speculative Decoding

1 code implementation27 Nov 2024 Ziyin Zhang, Jiahao Xu, Tian Liang, Xingyu Chen, Zhiwei He, Rui Wang, Zhaopeng Tu

Speculative Decoding (SD) has become an important technique in accelerating the inference speed of large language models.

8k

Identify Backdoored Model in Federated Learning via Individual Unlearning

1 code implementation1 Nov 2024 Jiahao Xu, Zikai Zhang, Rui Hu

Inspired by this, we propose MASA, a method that utilizes individual unlearning on local models to identify malicious models in FL.

Anomaly Detection Federated Learning +1

Fed-piLot: Optimizing LoRA Assignment for Efficient Federated Foundation Model Fine-Tuning

no code implementations14 Oct 2024 Zikai Zhang, Jiahao Xu, Ping Liu, Rui Hu

Specifically, Federated FMs (FedFMs) fine-tuning using low-rank adaptation (LoRA) modules instead of the full model over multiple clients can achieve both parameter efficiency and data privacy.

Federated Learning

Achieving Byzantine-Resilient Federated Learning via Layer-Adaptive Sparsified Model Aggregation

1 code implementation2 Sep 2024 Jiahao Xu, Zikai Zhang, Rui Hu

To address these challenges, we propose the Layer-Adaptive Sparsified Model Aggregation (LASA) approach, which combines pre-aggregation sparsification with layer-wise adaptive aggregation to improve robustness.

Federated Learning

Refuse Whenever You Feel Unsafe: Improving Safety in LLMs via Decoupled Refusal Training

2 code implementations12 Jul 2024 Youliang Yuan, Wenxiang Jiao, Wenxuan Wang, Jen-tse Huang, Jiahao Xu, Tian Liang, Pinjia He, Zhaopeng Tu

DeRTa incorporates two novel components: (1) Maximum Likelihood Estimation (MLE) with Harmful Response Prefix, which trains models to recognize and avoid unsafe content by appending a segment of harmful response to the beginning of a safe response, and (2) Reinforced Transition Optimization (RTO), which equips models with the ability to transition from potential harm to safety refusal consistently throughout the harmful response sequence.

Position

Learn from Heterophily: Heterophilous Information-enhanced Graph Neural Network

1 code implementation26 Mar 2024 Yilun Zheng, Jiahao Xu, Lihui Chen

Under circumstances of heterophily, where nodes with different labels tend to be connected based on semantic meanings, Graph Neural Networks (GNNs) often exhibit suboptimal performance.

Graph Learning Graph Neural Network +1

DistillCSE: Distilled Contrastive Learning for Sentence Embeddings

1 code implementation20 Oct 2023 Jiahao Xu, Wei Shao, Lihui Chen, Lemao Liu

This paper proposes the DistillCSE framework, which performs contrastive learning under the self-training paradigm with knowledge distillation.

Contrastive Learning Knowledge Distillation +2

Rethinking Cross-Domain Pedestrian Detection: A Background-Focused Distribution Alignment Framework for Instance-Free One-Stage Detectors

1 code implementation15 Sep 2023 Yancheng Cai, Bo Zhang, Baopu Li, Tao Chen, Hongliang Yan, Jingdong Zhang, Jiahao Xu

Therefore, we focus on cross-domain background feature alignment while minimizing the influence of foreground features on the cross-domain alignment stage.

Pedestrian Detection

SimCSE++: Improving Contrastive Learning for Sentence Embeddings from Two Perspectives

no code implementations22 May 2023 Jiahao Xu, Wei Shao, Lihui Chen, Lemao Liu

This paper improves contrastive learning for sentence embeddings from two perspectives: handling dropout noise and addressing feature corruption.

Contrastive Learning Sentence +1

Re-GAN: Data-Efficient GANs Training via Architectural Reconfiguration

1 code implementation CVPR 2023 Divya Saxena, Jiannong Cao, Jiahao Xu, Tarun Kulshrestha

Re-GAN stabilizes the GANs models with less data and offers an alternative to the existing GANs tickets and progressive growing methods.

Image Generation

Modulation and Classification of Mixed Signals Based on Deep Learning

no code implementations20 May 2022 Jiahao Xu, Zihuai Lin

Second, we investigate some deep learning models based on CNN (ResNet34, hierarchical structure) and other deep learning models (LSTM, CLDNN).

Classification Deep Learning

DisenE: Disentangling Knowledge Graph Embeddings

no code implementations28 Oct 2020 Xiaoyu Kou, Yankai Lin, Yuntao Li, Jiahao Xu, Peng Li, Jie zhou, Yan Zhang

Knowledge graph embedding (KGE), aiming to embed entities and relations into low-dimensional vectors, has attracted wide attention recently.

Entity Embeddings Knowledge Graph Embedding +2

LA-HCN: Label-based Attention for Hierarchical Multi-label TextClassification Neural Network

1 code implementation23 Sep 2020 Xinyi Zhang, Jiahao Xu, Charlie Soh, Lihui Chen

In this paper, we propose a Label-based Attention for Hierarchical Mutlti-label Text Classification Neural Network (LA-HCN), where the novel label-based attention module is designed to hierarchically extract important information from the text based on the labels from different hierarchy levels.

Multi Label Text Classification Multi-Label Text Classification +1

Cannot find the paper you are looking for? You can Submit a new open access paper.