Author name cluster

Shiyao Li

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.

5 papers

2 author rows

ICRA Conference 2025 Conference Paper

Knowledge-Driven Visual Target Navigation: Dual Graph Navigation

Shiyao Li
Ziyang Meng
Jiansong Pei
Jiahao Chen
Bingcheng Dong
Guangsheng Li
Shenglan Liu 0001
Feilong Wang

In unknown environments, navigating a robot by a given image to a specific location or instance is critical and challenging. The existing end-to-end approaches require simultaneous implicit learning of multiple subtasks, and modular approaches depend on metric information. Both approaches face high computational demands, often leading to difficulties in real-time updates and limited generalization, making them challenging to implement on resource-constrained devices. To address these challenges, we propose Dual Graph Navigation (DGN), a knowledge-driven, lightweight image instance navigation framework. DGN builds an External Knowledge Graph (EKG) from small-scale datasets to capture prior object correlations, efficiently guiding target exploration. During exploration, DGN builds an Internal Knowledge Graph (IKG) using an instance-aware module, which records explored objects based on reachability relationships rather than precise metric information. The IKG dynamically updates the EKG, enhancing the robot's adaptability to the current environment. Together, they realize topological perception and reduce computational overhead. Furthermore, unlike approaches characterized by over-dependence between components, DGN employs a plug-and-play modular design that allows independent training and flexible replacement of functional modules, effectively enhancing generalization performance while reducing training and deployment costs. Experiments illustrate that DGN generalizes well in different simulation environments (AI2-THOR, Habitat), achieving state-of-the-art performance on the ProcTHOR-10K dataset. It is compatible with three distinct real-world robot platforms, including edge computing devices without CUDA support. It exhibits a decision-making speed of 3. 8 to 5. 5 times over baseline methods. Further details can be found on the project page: https://dogplanningloyo.github.io/DGN/.

Details

ICLR Conference 2025 Conference Paper

NextBestPath: Efficient 3D Mapping of Unseen Environments

Shiyao Li
Antoine Guédon
Clémentin Boittiaux
Shizhe Chen
Vincent Lepetit

This work addresses the problem of active 3D mapping, where an agent must find an efficient trajectory to exhaustively reconstruct a new scene. Previous approaches mainly predict the next best view near the agent's location, which is prone to getting stuck in local areas. Additionally, existing indoor datasets are insufficient due to limited geometric complexity and inaccurate ground truth meshes. To overcome these limitations, we introduce a novel dataset AiMDoom with a map generator for the Doom video game, enabling to better benchmark active 3D mapping in diverse indoor environments. Moreover, we propose a new method we call next-best-path (NBP), which predicts long-term goals rather than focusing solely on short-sighted views. The model jointly predicts accumulated surface coverage gains for long-term goals and obstacle maps, allowing it to efficiently plan optimal paths with a unified model. By leveraging online data collection, data augmentation and curriculum learning, NBP significantly outperforms state-of-the-art methods on both the existing MP3D dataset and our AiMDoom dataset, achieving more efficient mapping in indoor environments of varying complexity.

Details

ICLR Conference 2025 Conference Paper

ViDiT-Q: Efficient and Accurate Quantization of Diffusion Transformers for Image and Video Generation

Tianchen Zhao
Tongcheng Fang
Haofeng Huang
Rui Wan
Widyadewi Soedarmadji
Enshu Liu
Shiyao Li
Zinan Lin 0001

Diffusion transformers have demonstrated remarkable performance in visual generation tasks, such as generating realistic images or videos based on textual instructions. However, larger model sizes and multi-frame processing for video generation lead to increased computational and memory costs, posing challenges for practical deployment on edge devices. Post-Training Quantization (PTQ) is an effective method for reducing memory costs and computational complexity. When quantizing diffusion transformers, we find that existing quantization methods face challenges when applied to text-to-image and video tasks. To address these challenges, we begin by systematically analyzing the source of quantization error and conclude with the unique challenges posed by DiT quantization. Accordingly, we design an improved quantization scheme: ViDiT-Q (**V**ideo \& **I**mage **Di**ffusion **T**ransformer **Q**uantization), tailored specifically for DiT models. We validate the effectiveness of ViDiT-Q across a variety of text-to-image and video models, achieving W8A8 and W4A8 with negligible degradation in visual quality and metrics. Additionally, we implement efficient GPU kernels to achieve practical 2-2.5x memory optimization and a 1.4-1.7x end-to-end latency speedup.

Details

NeurIPS Conference 2024 Conference Paper

Can LLMs Learn by Teaching for Better Reasoning? A Preliminary Study

Xuefei Ning
Zifu Wang
Shiyao Li
Zinan Lin
Peiran Yao
Tianyu Fu
Matthew B. Blaschko
Guohao Dai

Teaching to improve student models (e. g. , knowledge distillation) is an extensively studied methodology in LLMs. However, in human education, teaching enhances not only the students but also the teachers by fostering more rigorous and clearer reasoning, as well as deeper knowledge building. We ask: Can LLMs also learn by teaching (LbT) for better reasoning? If the answer is yes, we can potentially unlock the possibility of continuously advancing the models without solely relying on human-produced data or stronger models. In this paper, we provide a preliminary exploration of this question. We show that LbT ideas can be incorporated into existing LLM training/prompting pipelines and bring improvements. Specifically, we design three methods, each mimicking one of the three levels of LbT: observing students' feedback, learning from the feedback, and learning iteratively, with the goal of improving answer accuracy without training or improving models' inherent capability with fine-tuning. We reveal some findings: (1) Teaching materials that make it easier for students to learn (via in-context learning) have clearer and more accurate logic; (2) Weak-to-strong generalization: LbT might help improve strong models by teaching weak models; (3) Diversity in students might help: teaching multiple students could be better than teaching a single student or the teacher alone. We hope that our exploration can inspire future research on LbT and, more broadly, the adoption of advanced education techniques to improve LLMs. The code and website are at https: //github. com/imagination-research/lbt and https: //sites. google. com/view/llm-learning-by-teaching.

PDF Details DOI

ICML Conference 2024 Conference Paper

Evaluating Quantized Large Language Models

Shiyao Li
Xuefei Ning
Luning Wang
Tengxuan Liu
Xiangsheng Shi
Shengen Yan
Guohao Dai 0001
Huazhong Yang

Post-training quantization (PTQ) has emerged as a promising technique to reduce the cost of large language models (LLMs). Specifically, PTQ can effectively mitigate memory consumption and reduce computational overhead in LLMs. To meet the requirements of both high efficiency and performance across diverse scenarios, a comprehensive evaluation of quantized LLMs is essential to guide the selection of quantization methods. This paper presents a thorough evaluation of these factors by evaluating the effect of PTQ on Weight, Activation, and KV Cache on 11 model families, including OPT, LLaMA2, Falcon, Bloomz, Mistral, ChatGLM, Vicuna, LongChat, StableLM, Gemma, and Mamba, with parameters ranging from 125M to 180B. The evaluation encompasses five types of tasks: basic NLP, emergent ability, trustworthiness, dialogue, and long-context tasks. Moreover, we also evaluate the state-of-the-art (SOTA) quantization methods to demonstrate their applicability. Based on the extensive experiments, we systematically summarize the effect of quantization, provide recommendations to apply quantization techniques, and point out future directions. The code can be found in https: //github. com/thu-nics/qllm-eval.

Details