Author name cluster

Zhenzhong Lan

Possible papers associated with this exact author name in Arrow. This page groups case-insensitive exact name matches and is not a full identity disambiguation profile.

6 papers

2 author rows

AAMAS Conference 2026 Conference Paper

SocraticAgent: An Autonomous Agent for Unlocking Latent Knowledge in LLMs

Yang Yan
Yu Lu
Renjun Xu
Zhenzhong Lan

Reasoning failures in Large Language Models (LLMs) used by autonomous agents are often attributed to knowledge deficits, leading to a reliance on solutions like Retrieval-Augmented Generation (RAG) or parametric fine-tuning. This paper empirically demonstrates that this assumption is often flawed. We identify a quantifiable "knowledge recall gap": while modern LLMs possess 90-97% of the necessary facts for a task, they spontaneously apply only 57- 64% of this knowledge during reasoning. This reveals a significant performance gap rooted in a failure of recall, not a fundamental absence of knowledge. To address this, we introduce SocraticAgent, a zero-shot autonomous agent that emulates Socratic inquiry by guidinganLLMtofirstdeconstructaproblemandcomprehensively detail the internal knowledge required for its solution. Through a deterministic two-action cycle of (1) knowledge deconstruction and (2) grounded reasoning, it procedurally closes this recall gap without any model updates. Across a diverse suite of LLMs, SocraticAgent significantly improves reasoning accuracy, outperforming standard prompting and noisy external retrieval. Critically, our agentic, process-driven approach achieves performance competitive with expensive, data-dependent fine-tuning methods, but does so at inference time without any parametric changes. Our work demonstrates that a deliberative agentic process can serve as a powerful substitute for parametric memory adaptation. This paves the way for adaptable, capable autonomous reasoning systems, positioningagent-drivendeliberationasakeymechanismforunlocking latent knowledge within LLMs. Code and prompts are available at https: //github. com/kuri-leo/BigFive-LLM-Predictor.

PDF

NeurIPS Conference 2024 Conference Paper

AgentBoard: An Analytical Evaluation Board of Multi-turn LLM Agents

Chang Ma
Junlei Zhang
Zhihao Zhu
Cheng Yang
Yujiu Yang
Yaohui Jin
Zhenzhong Lan
Lingpeng Kong

Evaluating large language models (LLMs) as general-purpose agents is essential for understanding their capabilities and facilitating their integration into practical applications. However, the evaluation process presents substantial challenges. A primary obstacle is the benchmarking of agent performance across diverse scenarios within a unified framework, especially in maintaining partially-observable environments and ensuring multi-round interactions. Moreover, current evaluation frameworks mostly focus on the final success rate, revealing few insights during the process and failing to provide a deep understanding of the model abilities. To address these challenges, we introduce AgentBoard, a pioneering comprehensive benchmark and accompanied open-source evaluation framework tailored to analytical evaluation of LLM agents. AgentBoard offers a fine-grained progress rate metric that captures incremental advancements as well as a comprehensive evaluation toolkit that features easy assessment of agents for multi-faceted analysis through interactive visualization. This not only sheds light on the capabilities and limitations of LLM agents but also propels the interpretability of their performance to the forefront. Ultimately, AgentBoard serves as a significant step towards demystifying agent behaviors and accelerating the development of stronger LLM agents.

PDF Details DOI

AAAI Conference 2023 Conference Paper

Instance Smoothed Contrastive Learning for Unsupervised Sentence Embedding

Hongliang He
Junlei Zhang
Zhenzhong Lan
Yue Zhang

Contrastive learning-based methods, such as unsup-SimCSE, have achieved state-of-the-art (SOTA) performances in learning unsupervised sentence embeddings. However, in previous studies, each embedding used for contrastive learning only derived from one sentence instance, and we call these embeddings instance-level embeddings. In other words, each embedding is regarded as a unique class of its own, which may hurt the generalization performance. In this study, we propose IS-CSE (instance smoothing contrastive sentence embedding) to smooth the boundaries of embeddings in the feature space. Specifically, we retrieve embeddings from a dynamic memory buffer according to the semantic similarity to get a positive embedding group. Then embeddings in the group are aggregated by a self-attention operation to produce a smoothed instance embedding for further analysis. We evaluate our method on standard semantic text similarity (STS) tasks and achieve an average of 78.30%, 79.47%, 77.73%, and 79.42% Spearman’s correlation on the base of BERT-base, BERT-large, RoBERTa-base, and RoBERTa-large respectively, a 2.05%, 1.06%, 1.16% and 0.52% improvement compared to unsup-SimCSE.

PDF Details DOI

NeurIPS Conference 2021 Conference Paper

Dynamic Resolution Network

Mingjian Zhu
Kai Han
Enhua Wu
Qiulin Zhang
Ying Nie
Zhenzhong Lan
Yunhe Wang

Deep convolutional neural networks (CNNs) are often of sophisticated design with numerous learnable parameters for the accuracy reason. To alleviate the expensive costs of deploying them on mobile devices, recent works have made huge efforts for excavating redundancy in pre-defined architectures. Nevertheless, the redundancy on the input resolution of modern CNNs has not been fully investigated, i. e. , the resolution of input image is fixed. In this paper, we observe that the smallest resolution for accurately predicting the given image is different using the same neural network. To this end, we propose a novel dynamic-resolution network (DRNet) in which the input resolution is determined dynamically based on each input sample. Wherein, a resolution predictor with negligible computational costs is explored and optimized jointly with the desired network. Specifically, the predictor learns the smallest resolution that can retain and even exceed the original recognition accuracy for each image. During the inference, each input image will be resized to its predicted resolution for minimizing the overall computation burden. We then conduct extensive experiments on several benchmark networks and datasets. The results show that our DRNet can be embedded in any off-the-shelf network architecture to obtain a considerable reduction in computational complexity. For instance, DR-ResNet-50 achieves similar performance with an about 34% computation reduction, while gaining 1. 4% accuracy increase with 10% computation reduction compared to the original ResNet-50 on ImageNet. Code will be available at https: //gitee. com/mindspore/models/tree/master/research/cv/DRNet.

PDF Details

ICLR Conference 2020 Conference Paper

ALBERT: A Lite BERT for Self-supervised Learning of Language Representations

Zhenzhong Lan
Mingda Chen
Sebastian Goodman
Kevin Gimpel
Piyush Sharma
Radu Soricut

Increasing model size when pretraining natural language representations often results in improved performance on downstream tasks. However, at some point further model increases become harder due to GPU/TPU memory limitations and longer training times. To address these problems, we present two parameter-reduction techniques to lower memory consumption and increase the training speed of BERT~\citep{devlin2018bert}. Comprehensive empirical evidence shows that our proposed methods lead to models that scale much better compared to the original BERT. We also use a self-supervised loss that focuses on modeling inter-sentence coherence, and show it consistently helps downstream tasks with multi-sentence inputs. As a result, our best model establishes new state-of-the-art results on the GLUE, RACE, and \squad benchmarks while having fewer parameters compared to BERT-large. The code and the pretrained models are available at https://github.com/google-research/ALBERT.

Details

NeurIPS Conference 2014 Conference Paper

Self-Paced Learning with Diversity

Lu Jiang
Deyu Meng
Shoou-I Yu
Zhenzhong Lan
Shiguang Shan
Alexander Hauptmann

Self-paced learning (SPL) is a recently proposed learning regime inspired by the learning process of humans and animals that gradually incorporates easy to more complex samples into training. Existing methods are limited in that they ignore an important aspect in learning: diversity. To incorporate this information, we propose an approach called self-paced learning with diversity (SPLD) which formalizes the preference for both easy and diverse samples into a general regularizer. This regularization term is independent of the learning objective, and thus can be easily generalized into various learning tasks. Albeit non-convex, the optimization of the variables included in this SPLD regularization term for sample selection can be globally solved in linearithmic time. We demonstrate that our method significantly outperforms the conventional SPL on three real-world datasets. Specifically, SPLD achieves the best MAP so far reported in literature on the Hollywood2 and Olympic Sports datasets.

PDF Details