https://github.com/jacobmarks/awesome-neurips-2023
Conference schedule, top papers, and analysis of the data for NeurIPS 2023!
https://github.com/jacobmarks/awesome-neurips-2023
List: awesome-neurips-2023
ai awesome awesome-list awesome-readme machine-learning neurips neurips-2023
Last synced: 4 months ago
JSON representation
Conference schedule, top papers, and analysis of the data for NeurIPS 2023!
- Host: GitHub
- URL: https://github.com/jacobmarks/awesome-neurips-2023
- Owner: jacobmarks
- Created: 2023-12-05T08:18:19.000Z (about 1 year ago)
- Default Branch: main
- Last Pushed: 2023-12-15T21:41:04.000Z (about 1 year ago)
- Last Synced: 2024-10-26T06:52:16.857Z (4 months ago)
- Topics: ai, awesome, awesome-list, awesome-readme, machine-learning, neurips, neurips-2023
- Language: Jupyter Notebook
- Homepage:
- Size: 11.9 MB
- Stars: 106
- Watchers: 2
- Forks: 7
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
- ultimate-awesome - awesome-neurips-2023 - Conference schedule, top papers, and analysis of the data for NeurIPS 2023!. (Other Lists / Julia Lists)
README
# Awesome NeurIPS 2023 Info
Caption: Wordcloud of all NeurIPS 2023 titles
Welcome to the hub for all things [NeurIPS 2023](https://neurips.cc/)! We scraped the data for all 3500+ NeurIPS projects and dove into the depths of Hugging Face, GitHub, LinkedIn, and Arxiv to pick out the most interesting content.
In this repo, you will find:
- [Data Analysis](#data-analysis): detailed analysis of the titles and abstracts from NeurIPS 2023 accepted papers
- [Awesome Projects](#cool-neurips-projects): synthesized collection of 40 NeurIPS 2023 papers you won't want to miss
- [Conference Schedule](#conference-schedule): comprehensive listing of all NeurIPS 2023 projects (title, authors, abstract) organized by poster session and sorted alphabetically
## Data Analysis
The raw data is included in this repo. If you have ideas for other interesting analyses, feel free to create an issue or submit a PR!
For now, insights are organized into the following categories:
- Authors
- Titles
- Abstracts
🔍 For the data analysis itself, check out the [Jupyter Notebook](./analysis.ipynb)!
🔍 And check out the blog post synthesizing the results [here](https://voxel51.com/blog/neurips-2023-and-the-state-of-ai-research/).
Authors
#### Most prolific authors
The top 10 authors with the most papers at NeurIPS 2023 are:
- Bo Li: 15 papers
- Ludwig Schmidt: 14 papers
- Bo Han: 13 papers
- Mihaela van der Schaar: 13 papers
- Hao Wang: 12 papers
- Dacheng Tao: 11 papers
- Bernhard Schölkopf: 11 papers
- Masashi Sugiyama: 11 papers
- Andreas Krause: 11 papers
- Tongliang Liu: 11 papers
#### Number of unique authors
There were 13,012 unique authors at NeurIPS 2023, up from 9913 at NeurIPS 2022.
This continues the exponential explosion of unique authors over the past decade.
#### Number of authors per paper
- The average number of authors per paper was **4.98**, up from 4.66 at NeurIPS 2022.
- Additionally, there were a handful of single-author papers, in contrast to NeurIPS 2022, where the minimum number of authors was 2.
- The paper with the most authors was [ClimSim: A large multi-scale dataset for hybrid physics-ML climate emulation](https://arxiv.org/abs/2306.08754)
Titles
#### Title Length
- The average title length was **8.72** words, up from 8.48 at NeurIPS 2022. This
continues an ongoing trend of title lengthening:
#### Prevalence of Acronyms
22% of titles introduced an acronym, up from 18% at NeurIPS 2022.
#### LaTeX in Titles
- 1.3% of titles contained LaTeX, whereas none of the titles at NeurIPS 2022 contained LaTeX.
Abstracts
#### Abstract Length
- The longest abstract was from [[Re] On the Reproducibility of FairCal: Fairness Calibration for Face Verification](https://neurips.cc/virtual/2023/poster/74168), which has 373 words.
- The shortest abstract was from [Improved Convergence in High Probability of Clipped Gradient Methods with Heavy Tailed Noise](https://neurips.cc/virtual/2023/poster/70813), which has 29 words.
#### GitHub Reigns Supreme
- Out of the 3581 abstracts, 675 explicitly mention GitHub, including a link to their code, models, or data.
- Only 79 abstracts include a URL that is *not* GitHub.
#### Modalities, Models, and Tasks
Using a [CLIP](https://github.com/openai/CLIP) model, we zero-shot
classified/predicted the modality of focus for each paper based on its abstract.
The categories were `["vision", "text", "audio", "tabular", "time series", "multimodal"]`.
By far the biggest category was multimodal, with a count of 1296. However, the
CLIP model's inclination towards "multimodal" may be somewhat biased by trying
to partially fit other modalities — the words `multi-modal` and `multimodal` only
show up in 156 abstracts, and phrases like `vision-language` and `text-to-image`
only appear a handful of times across the dataset.
Themes occurring frequently include:
- "benchmark": 730
- ("generation", "generate"): 681
- ("efficient", "efficiency"): 963
- "agent": 280
- ("llm", "large language model"): 238
## Cool NeurIPS Projects
| **Title** | **Paper** | **Code** | **Project Page** | **Hugging Face** | **Blog** |
|:---------:|:---------:|:--------:|:----------------:|:----------------:|:--------:|
| An Inverse Scaling Law for CLIP Training | [](https://arxiv.org/abs/2305.07017) | [](https://github.com/UCSC-VLAA/CLIPA)| | | |
| Augmenting Language Models with Long-Term Memory | [](https://arxiv.org/abs/2306.07174) | [](https://github.com/Victorwz/LongMem)| | | |
| Chameleon: Plug-and-Play Compositional Reasoning with Large Language Models | [](https://arxiv.org/abs/2304.09842) | [](https://github.com/lupantech/chameleon-llm)| [Project](https://chameleon-llm.github.io/) | | [Blog](https://medium.com/voxel51/neurips-2023-survival-guide-2f957d5b07c9#b961) |
| Cheap and Quick: Efficient Vision-Language Instruction Tuning for Large Language Models | [](https://arxiv.org/abs/2305.15023) | [](https://github.com/luogen1996/LaVIN)| [Project](https://luogen1996.github.io/lavin/) | | [Blog](https://medium.com/voxel51/neurips-2023-survival-guide-2f957d5b07c9#9dcb) |
| DataComp: In search of the next generation of multimodal datasets | [](https://arxiv.org/abs/2304.14108) | [](https://github.com/mlfoundations/datacomp)| [Project](https://www.datacomp.ai/) | | [Blog](https://medium.com/voxel51/neurips-2023-survival-guide-2f957d5b07c9#1ede) |
| Direct Preference Optimization: Your Language Model is Secretly a Reward Model | [](https://arxiv.org/abs/2305.18290) | [](https://github.com/eric-mitchell/direct-preference-optimization)| | | [Blog](https://medium.com/@joaolages/direct-preference-optimization-dpo-622fc1f18707) |
| DreamSim: Learning New Dimensions of Human Visual Similarity using Synthetic Data | [](https://arxiv.org/abs/2306.09344) | [](https://github.com/ssundaram21/dreamsim)| [Project](https://dreamsim-nights.github.io/) | | [Blog](https://medium.com/voxel51/teaching-androids-to-dream-of-sheep-18d72f44f2b) |
| Fine-Tuning Language Models with Just Forward Passes | [](https://arxiv.org/abs/2305.17333) | [](https://github.com/princeton-nlp/MeZO)| | | [Blog](https://gaotianyu.xyz/blog/2023/11/14/mezo/) |
| Generating Images with Multimodal Language Models | [](https://arxiv.org/abs/2305.17216) | [](https://github.com/kohjingyu/gill)| [Project](https://jykoh.com/gill) | | |
| Holistic Evaluation of Text-To-Image Models | [](https://arxiv.org/abs/2311.04287) | [](https://github.com/stanford-crfm/heim)| [Project](https://crfm.stanford.edu/heim/latest/) | | [Blog](https://medium.com/voxel51/neurips-2023-survival-guide-2f957d5b07c9#b237) |
| HuggingGPT: Solving AI Tasks with ChatGPT and its Friends in Hugging Face | [](https://arxiv.org/abs/2303.17580) | [](https://github.com/microsoft/JARVIS)| | [](https://huggingface.co/spaces/microsoft/HuggingGPT) | |
| ImageReward: Learning and Evaluating Human Preferences for Text-to-Image Generation | [](https://arxiv.org/abs/2304.05977) | [](https://github.com/THUDM/ImageReward)| | [](https://huggingface.co/THUDM/ImageReward) | [Blog](https://medium.com/voxel51/neurips-2023-survival-guide-2f957d5b07c9#ImageReward) |
| InstructBLIP: Towards General-purpose Vision-Language Models with Instruction Tuning | [](https://arxiv.org/abs/2305.06500) | [](https://github.com/salesforce/LAVIS/tree/main/projects/instructblip)| | [](Salesforce/instructblip-vicuna-7b) | [Blog](https://medium.com/voxel51/neurips-2023-survival-guide-2f957d5b07c9#b689) |
| Judging LLM-as-a-Judge with MT-Bench and Chatbot Arena | [](https://arxiv.org/abs/2306.05685) | [](https://github.com/lm-sys/FastChat/tree/main/fastchat/llm_judge)| | | |
| LAMM: Multi-Modal Large Language Models and Applications as AI Agents | [](https://arxiv.org/abs/2306.06687) | [](https://github.com/OpenGVLab/LAMM)| [Project](https://openlamm.github.io/) | [](https://medium.com/voxel51/neurips-2023-survival-guide-2f957d5b07c9#a998) | |
| LIMA: Less Is More for Alignment | [](https://arxiv.org/abs/2305.11206) | | | | [Blog](https://medium.com/version-1/comprehensive-analysis-of-lima-less-is-more-for-alignment-8967687ea432) |
| LLM-Pruner: On the Structural Pruning of Large Language Models | [](https://arxiv.org/abs/2305.11627) | [](https://github.com/horseee/LLM-Pruner)| | | |
| LightZero: A Unified Benchmark for Monte Carlo Tree Search in General Sequential Decision Scenario | [](https://arxiv.org/abs/2310.08348) | [](https://github.com/opendilab/LightZero)| | | |
| MVDiffusion: Enabling Holistic Multi-view Image Generation with Correspondence-Aware Diffusion | [](https://arxiv.org/abs/2307.01097) | [](https://github.com/Tangshitao/MVDiffusion)| [Project](https://mvdiffusion.github.io/) | | |
| MagicBrush: A Manually Annotated Dataset for Instruction-Guided Image Editing | [](https://arxiv.org/abs/2306.10012) | [](https://github.com/OSU-NLP-Group/MagicBrush)| [Project](https://osu-nlp-group.github.io/MagicBrush/) | [](https://huggingface.co/datasets/osunlp/MagicBrush) | [Blog](https://medium.com/voxel51/neurips-2023-survival-guide-2f957d5b07c9#MagicBrush) |
| Mathematical Capabilities of ChatGPT | [](https://arxiv.org/abs/2301.13867) | [](https://github.com/friederrr/GHOSTS)| [Project](https://ghosts.friederrr.org/) | | |
| Michelangelo: Conditional 3D Shape Generation based on Shape-Image-Text Aligned Latent Representation | [](https://arxiv.org/abs/2306.17115) | [](https://github.com/NeuralCarver/Michelangelo)| [Project](https://neuralcarver.github.io/michelangelo/) | | |
| Motion-X: A Large-scale 3D Expressive Whole-body Human Motion Dataset | [](https://arxiv.org/abs/2307.00818) | [](https://github.com/IDEA-Research/Motion-X)| [Project](https://motion-x-dataset.github.io/) | | |
| MotionGPT: Human Motion as Foreign Language | [](https://arxiv.org/abs/2306.14795) | [](https://github.com/OpenMotionLab/MotionGPT)| [Project](https://motion-gpt.github.io/) | [](https://huggingface.co/spaces/OpenMotionLab/MotionGPT) | [Blog](https://medium.com/@kaveh.kamali/unlocking-the-language-of-motion-meet-motiongpt-8efee7700fd3) |
| OBELICS: An Open Web-Scale Filtered Dataset of Interleaved Image-Text Documents | [](https://arxiv.org/abs/2306.16527) | [](https://github.com/huggingface/OBELICS)| | [](https://huggingface.co/datasets/HuggingFaceM4/OBELICS) | [Blog](https://medium.com/voxel51/neurips-2023-survival-guide-2f957d5b07c9#e062) |
| Photoswap: Personalized Subject Swapping in Images | [](https://arxiv.org/abs/2305.18286) | [](https://github.com/eric-ai-lab/photoswap)| [Project](https://photoswap.github.io/) | | |
| Pick-a-Pic: An Open Dataset of User Preferences for Text-to-Image Generation | [](https://arxiv.org/abs/2305.01569) | [](https://github.com/yuvalkirstain/PickScore)| | [](https://huggingface.co/datasets/yuvalkirstain/pickapic_v1) | [Blog](https://medium.com/voxel51/neurips-2023-survival-guide-2f957d5b07c9#3a8f) |
| QLoRA: Efficient Finetuning of Quantized LLMs | [](https://arxiv.org/abs/2305.14314) | [](https://github.com/artidoro/qlora)| | [](https://huggingface.co/spaces/uwnlp/guanaco-playground-tgi) | [Blog](https://medium.com/@dariussingh/qlora-a-new-way-to-finetune-llms-4a5ff292903d) |
| Reflexion: Language Agents with Verbal Reinforcement Learning | [](https://arxiv.org/abs/2303.11366) | [](https://github.com/noahshinn/reflexion)| | | [Blog](https://nanothoughts.substack.com/p/reflecting-on-reflexion) |
| ResShift: Efficient Diffusion Model for Image Super-resolution by Residual Shifting | [](https://arxiv.org/abs/2307.12348) | [](https://github.com/zsyOAOA/ResShift)| [Project](https://zsyoaoa.github.io/projects/resshift/) | | [Blog](https://www.marktechpost.com/2023/08/01/ntu-singapore-researchers-introduce-resshift-a-new-upscaler-model-that-uses-residual-shifting-and-achieves-image-super-resolution-faster-compared-to-other-methods/) |
| Segment Anything in 3D with NeRFs | [](https://arxiv.org/abs/2304.12308) | [](https://github.com/Jumpat/SegmentAnythingin3D)| [Project](https://jumpat.github.io/SA3D/) | | [Blog](https://www.marktechpost.com/2023/05/22/when-sam-meets-nerf-this-ai-model-can-segment-anything-in-3d/) |
| Segment Anything in High Quality | [](https://arxiv.org/abs/2306.01567) | [](https://github.com/SysCV/sam-hq)| | [](https://huggingface.co/spaces/sam-hq-team/sam-hq) | [Blog](https://supervisely.com/blog/segment-anything-in-high-quality-HQ-SAM/) |
| Segment Everything Everywhere All at Once | [](https://arxiv.org/abs/2304.06718) | [](https://github.com/UX-Decoder/Segment-Everything-Everywhere-All-At-Once)| | | |
| Self-Refine: Iterative Refinement with Self-Feedback | [](https://arxiv.org/abs/2303.17651) | [](https://github.com/madaan/self-refine)| [Project](https://selfrefine.info/) | | [Blog](https://www.marktechpost.com/2023/04/07/this-ai-paper-introduce-self-refine-a-framework-for-improving-initial-outputs-from-llms-through-iterative-feedback-and-refinement/) |
| Simple and Controllable Music Generation | [](https://arxiv.org/abs/2306.05284) | [](https://github.com/facebookresearch/audiocraft)| | | [Blog](https://vivek-murali.medium.com/simple-and-controllable-music-generation-musicgen-by-meta-cc8863d73792) |
| Squeeze, Recover and Relabel: Dataset Condensation at ImageNet Scale From A New Perspective | [](https://arxiv.org/abs/2306.13092) | [](https://github.com/VILA-Lab/SRe2L)| | | |
| The RefinedWeb Dataset for Falcon LLM: Outperforming Curated Corpora with Web Data, and Web Data Only | [](https://arxiv.org/abs/2306.01116) | | | [](https://huggingface.co/datasets/tiiuae/falcon-refinedweb) | [Blog](https://medium.com/dair-ai/papers-explained-59-falcon-26831087247f) |
| Toolformer: Language Models Can Teach Themselves to Use Tools | [](https://arxiv.org/abs/2302.04761) | | | | [Blog](https://kikaben.com/toolformer-2023/) |
| Unlimiformer: Long-Range Transformers with Unlimited Length Input | [](https://arxiv.org/abs/2305.01625) | [](https://github.com/abertsch72/unlimiformer)| | | [Blog](https://pub.towardsai.net/unlimiformer-long-range-transformers-with-unlimited-length-input-3725f69b0d03?gi=86f7474233cb) |
| Visual Instruction Tuning | [](https://arxiv.org/abs/2304.08485) | [](https://github.com/haotian-liu/LLaVA)| [Project](https://llava-vl.github.io/) | [](https://huggingface.co/spaces/badayvedat/LLaVA) | [Blog](https://medium.com/voxel51/understanding-llava-large-language-and-vision-assistant-8b7772f5eec4) |
## Conference Schedule
Note: GitHub automatically truncates files larger than 512 KB. To have all papers display on GitHub, we've split the file up by session.
[Poster Session 1](schedule/Poster_Session1.md)
[Poster Session 2](schedule/Poster_Session2.md)
[Poster Session 3](schedule/Poster_Session3.md)
[Poster Session 4](schedule/Poster_Session4.md)
[Poster Session 5](schedule/Poster_Session5.md)
[Poster Session 6](schedule/Poster_Session6.md)
[Posters Not Presented](schedule/Not_Presented_Posters.md)