{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/ella-equip-diffusion-models-with-llm-for","title":"ELLA: Equip Diffusion Models with LLM for Enhanced Semantic Alignment","arxiv_id":"2403.05135","date":"2024-03-08","proceeding":null,"authors":["XiWei Hu","Rui Wang","Yixiao Fang","Bin Fu","Pei Cheng","Gang Yu"],"abstract":"Diffusion models have demonstrated remarkable performance in the domain of text-to-image generation. However, most widely used models still employ CLIP as their text encoder, which constrains their ability to comprehend dense prompts, encompassing multiple objects, detailed attributes, complex relationships, long-text alignment, etc. In this paper, we introduce an Efficient Large Language Model Adapter, termed ELLA, which equips text-to-image diffusion models with powerful Large Language Models (LLM) to enhance text alignment without training of either U-Net or LLM. To seamlessly bridge two pre-trained models, we investigate a range of semantic alignment connector designs and propose a novel module, the Timestep-Aware Semantic Connector (TSC), which dynamically extracts timestep-dependent conditions from LLM. Our approach adapts semantic features at different stages of the denoising process, assisting diffusion models in interpreting lengthy and intricate prompts over sampling timesteps. Additionally, ELLA can be readily incorporated with community models and tools to improve their prompt-following capabilities. To assess text-to-image models in dense prompt following, we introduce Dense Prompt Graph Benchmark (DPG-Bench), a challenging benchmark consisting of 1K dense prompts. Extensive experiments demonstrate the superiority of ELLA in dense prompt following compared to state-of-the-art methods, particularly in multiple object compositions involving diverse attributes and relationships.","url_abs":"https://arxiv.org/abs/2403.05135v1","url_pdf":"https://arxiv.org/pdf/2403.05135v1.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"ella-equip-diffusion-models-with-llm-for","repo_url":"https://github.com/shihaozhaozsh/lavi-bridge","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"MIT"}},{"paper_slug":"ella-equip-diffusion-models-with-llm-for","repo_url":"https://github.com/tencentqqgylab/ella","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"denoising","task_name":"Denoising"},{"task_slug":"image-generation","task_name":"Image Generation"},{"task_slug":"language-modelling","task_name":"Language Modelling"},{"task_slug":"large-language-model","task_name":"Large Language Model"},{"task_slug":"text-to-image-generation-1","task_name":"Text to Image Generation"},{"task_slug":"text-to-image-generation","task_name":"Text-to-Image Generation"}],"methods":[{"method_slug":"adapter","method_name":"Adapter"},{"method_slug":"clip","method_name":"CLIP"},{"method_slug":"concatenated-skip-connection","method_name":"Concatenated Skip Connection"},{"method_slug":"convolution","method_name":"Convolution"},{"method_slug":"diffusion","method_name":"Diffusion"},{"method_slug":"max-pooling","method_name":"Max Pooling"},{"method_slug":"relu","method_name":"ReLU"},{"method_slug":"u-net","method_name":"U-Net"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2403.05135","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2403.05135"}},"developers":"https://syntology.ai/developers","read_at":"2026-09-24T18:15:14+00:00","read_at_is":"when the build read Syntology's graph, not when any sample ran","claim":"Per-sample execution status on synthesized fixtures; not a correctness claim about the paper. Samples come from repositories linked to the paper, official or community; repo_kind says which.","repos":[{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/shihaozhaozsh/lavi-bridge","reach":{"status":"ok","spdx":"MIT"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/tencentqqgylab/ella","reach":null}],"summary":{"ran_draft_wrong":3,"ran_honours":2},"by_repo_kind":{"listed":{"samples":5,"ran":5,"repositories":1}},"repo_kind_vocabulary":{"official":"The archive marks this repository official for the paper","named_in_paper":"The archive records that the paper mentions this repository; it is not marked official","listed":"In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper","found_in_text":"Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted","community":"Not in the archive's code links for this paper; a community repository Syntology harvested"},"n_pointer_only_for_licence":0,"samples":[{"code_sha256_prefix":"d7c600805e21bbc6","entry":"compute_dpg_one_sample","repo":"tencentqqgylab/ella","repo_kind":"listed","path":"dpg_bench/compute_dpg_bench.py","file_url":"https://github.com/tencentqqgylab/ella/blob/HEAD/dpg_bench/compute_dpg_bench.py","link_basis":"first_harvest_node","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"OUTPUT_MISDECLARED","metamorphic_tier":"deterministic","behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"d7c600805e21bbc6"}},{"code_sha256_prefix":"10dc0d8ce3bf1721","entry":"crop_image","repo":"tencentqqgylab/ella","repo_kind":"listed","path":"dpg_bench/compute_dpg_bench.py","file_url":"https://github.com/tencentqqgylab/ella/blob/HEAD/dpg_bench/compute_dpg_bench.py","link_basis":"first_harvest_node","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"MISDECLARED","metamorphic_tier":"well_formed","behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"10dc0d8ce3bf1721"}},{"code_sha256_prefix":"90e281d5839c51db","entry":"generate_image_with_fixed_max_length","repo":"tencentqqgylab/ella","repo_kind":"listed","path":"inference.py","file_url":"https://github.com/tencentqqgylab/ella/blob/HEAD/inference.py","link_basis":"first_harvest_node","language":"python","status":"ran_honours","verification_level":1,"contract_check":"HONOURS","metamorphic_tier":"well_formed","behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"90e281d5839c51db"}},{"code_sha256_prefix":"c9b4fcb3221db452","entry":"generate_image_with_flexible_max_length","repo":"tencentqqgylab/ella","repo_kind":"listed","path":"inference.py","file_url":"https://github.com/tencentqqgylab/ella/blob/HEAD/inference.py","link_basis":"first_harvest_node","language":"python","status":"ran_honours","verification_level":1,"contract_check":"HONOURS","metamorphic_tier":"well_formed","behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"c9b4fcb3221db452"}},{"code_sha256_prefix":"6e653207d66d539a","entry":"prepare_dpg_data","repo":"tencentqqgylab/ella","repo_kind":"listed","path":"dpg_bench/compute_dpg_bench.py","file_url":"https://github.com/tencentqqgylab/ella/blob/HEAD/dpg_bench/compute_dpg_bench.py","link_basis":"first_harvest_node","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"OUTPUT_MISDECLARED","metamorphic_tier":"deterministic","behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"6e653207d66d539a"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}