Notice: The reproducibility variables underlying each score are classified using an automated LLM-based pipeline, validated against a manually labeled dataset. LLM-based classification introduces uncertainty and potential bias; scores should be interpreted as estimates. Full accuracy metrics and methodology are described in [1].

Expanding the Category of Classifiers with LLM Supervision

Authors: Derui Lyu, Xiangyu Wang, Taiyu Ban, Lyuzhou Chen, Xiren Zhou, Huanhuan Chen

IJCAI 2025 | Venue PDF | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Experimental results show that CEMIL outperforms existing methods using expert-constructed attributes, demonstrating its effectiveness for fully automated classifier expansion without human participation. Extensive experiments demonstrate that CEMIL, which operates solely with an LLM, consistently outperforms state-of-the-art ZSL methods that rely on expert-constructed attributes, across a variety of ZSL datasets.
Researcher Affiliation Academia Derui Lyu , Xiangyu Wang , Taiyu Ban , Lyuzhou Chen , Xiren Zhou , Huanhuan Chen University of Science and Technology of China EMAIL, EMAIL
Pseudocode No The paper describes the methodology using textual explanations and a framework diagram (Figure 2), but no structured pseudocode or algorithm blocks are explicitly provided.
Open Source Code No The paper does not contain any explicit statement about the release of source code for the described methodology, nor does it provide a link to a code repository.
Open Datasets Yes Methods are evaluated on three widely used datasets: 1) AWA2 [Xian et al., 2018], an animal classification dataset featuring 50 mammal species; 2) CUB [Wah et al., 2011], a dataset containing 200 bird species; and 3) SUN [Patterson et al., 2014], a scene recognition dataset with 717 categories.
Dataset Splits Yes Each dataset is accompanied by expert-constructed class attributes and is divided into seen and unseen classes based on the splitting scheme in [Xian et al., 2017]. We perform evaluations of the methods under both standard and generalized ZSL settings.
Hardware Specification Yes Experiments are conducted on an Nvidia Ge Force RTX 4090 24GB GPU.
Software Dependencies No The paper mentions using 'GPT-4o [Open AI, 2023] as the LLM and the text encoder of CLIP [Radford et al., 2021] as the embedding model' but does not specify versions for any other software dependencies, libraries, or programming languages used for implementation.
Experiment Setup Yes The initial expected view number is set to 50. Each encoder is a single-layer MLP, while each decoder is a two-layer MLP with a hidden dimension 4096. The dimensions of the attention vectors are 2048. Neural network parameters are initialized randomly from a standard normal distribution. The Adam optimizer is used for training, with up to 500 epochs and an early stopping strategy. The learning rate is set to 1e-5, with batch sizes configured as 10, 16, and 32 for AWA2, CUB, and SUN, respectively.