This was work done by an amazing team---co-led by Anne-Sofie Maerten and Juliane Verwiebe, w/ @ShyamgopalKart1, @AmyPr, Johan Wagemans, and Matthias Bethge at the Tübingen AI Center and KU Leuven.
Deutsch
Bethge Lab
324 posts
Bethge Lab
@bethgelab
Perceiving Neural Networks
Tübingen, Germany เข้าร่วม Temmuz 2017
243 กำลังติดตาม3.3K ผู้ติดตาม
🔎 Key finding: Users significantly preferred images optimized for their individual preferences (1065 ELO) compared to un-optimized baselines (1016), all while maintaining photorealism.
English
🚀New work from our group: Personalizing Text-to-Image Generation to Individual Taste! ✨
Paper: arxiv.org/abs/2604.07427
Project Page: pamela-bench.github.io
Dataset: huggingface.co/datasets/bethg…
🧵👇
English
Bethge Lab รีทวีตแล้ว
🚨We're hiring a Research Coordinator for our collaborative research center "Robust Vision" @uni_tue—a leadership role at the heart of a major @dfg_public consortium spanning neuroscience, ML & computer vision.
Requirements: MSc/PhD + science management experience. Apply by March 15 👇 join our amazing team with @mackelab @bethgelab & many more!
📷 uni-tuebingen.de/en/university/…
Please spread the word 🙏
English
Bethge Lab รีทวีตแล้ว
🎉Thrilled to share that CABS is accepted to #CVPR2026! See you in Denver!
Everything is now open-sourced:
Webpage:cabs-vlp.github.io
DataConcept:huggingface.co/datasets/bethg…
Paper:arxiv.org/abs/2511.20643
Code:github.com/bethgelab/cabs
Excited to see what the community builds!
Adhiraj Ghosh@adhiraj_ghosh98
🚨Current data curation results in the creation of static datasets and the use of model-based filters that induce many biases. Can we fix this? We propose ✨CABS✨, a flexible concept-aware online batch curation method that improves CLIP pretraining! arxiv.org/abs/2511.20643 🧵👇
English
Checkout our #NeurIPS2025 paper on combining deep networks with mechanistic models to improve visual behavior.
Where: Poster #2103, Session 4, Exhibit Hall C,D,E
When: 04/12 4:30PM PST
Go say hi to @fededagos!
Federico D'Agostino@fededagos
🚨 New paper at #NeurIPS2025! A systematic fixation-level comparison of a performance-optimized DNN scanpath model and a mechanistic cognitive model reveals behaviourally relevant mechanisms that can be added to the mechanistic model to substantially improve performance. 🧵👇
English
Hot off the press: new paper that demonstrates online data curation significantly outperforms CLIP and SigLIP pretraining. It also provides empericial evidence for specific curation for classification and retrieval.
Check it out below👇
Adhiraj Ghosh@adhiraj_ghosh98
🚨Current data curation results in the creation of static datasets and the use of model-based filters that induce many biases. Can we fix this? We propose ✨CABS✨, a flexible concept-aware online batch curation method that improves CLIP pretraining! arxiv.org/abs/2511.20643 🧵👇
English
New work from our lab conducting a critical audit of a recent work introducing spatial supersensing!
Check it out below👇
Vishaal Udandarao@vishaal_urao
🚀 New paper! arxiv.org/abs/2511.16655 Recently, Cambrian-S released models & two benchmarks (VSR & VSC) for “spatial supersensing” in video! We found: 1️⃣ Simple no-frame baseline (NoSense) ~perfectly solves VSR! 2️⃣ Tiny sanity check collapses Cambrian-S perf to 0% on VSC! 🧵👇
English
Bethge Lab รีทวีตแล้ว
Bethge Lab รีทวีตแล้ว
🚀New Paper
arxiv.org/abs/2510.20860
We conduct a systematic data-centric study for speech-language pretraining, to improve end-to-end spoken-QA! 🎙️🤖
Using our data-centric insights, we pretrain a 3.8B SpeechLM (called SpeLangy) outperforming 3x larger models!
🧵👇
English
Our Sober Reasoning paper (accepted at COLM'25) was recently featured on the State of AI Report 2025!!
Nathan Benaich@nathanbenaich
🪩The one and only @stateofai 2025 is live! 🪩 It’s been a monumental 12 months for AI. Our 8th annual report is the most comprehensive it's ever been, covering what you *need* to know about research, industry, politics, safety and our new usage data. My highlight reel:
English
Bethge Lab รีทวีตแล้ว
Great to see that our work is featured in the @stateofai 2025 🎉
@hrdkbhatnagar @vishaal_urao @SamuelAlbanie @AmyPrb @MatthiasBethge
Nathan Benaich@nathanbenaich
🪩The one and only @stateofai 2025 is live! 🪩 It’s been a monumental 12 months for AI. Our 8th annual report is the most comprehensive it's ever been, covering what you *need* to know about research, industry, politics, safety and our new usage data. My highlight reel:
English
Bethge Lab รีทวีตแล้ว
Bethge Lab รีทวีตแล้ว
Do language models have algorithmic creativity?
To find out, we built AlgoTune, a benchmark challenging agents to optimize 100+ algorithms like gzip compression, AES encryption and PCA. Frontier models struggle, finding only surface-level wins. Lots of headroom here!🧵⬇️
English
🧠🤖 We’re hiring a Postdoc in NeuroAI!
Join CRC1233 "Robust Vision" (Uni Tübingen) to build benchmarks & evaluation methods for vision models, bridging brain & AI. Work with top faculty & shape vision research.
Apply: tinyurl.com/3jtb4an6
#NeuroAI #Jobs
English
Recent work from our lab trying to ask questions on how to fairly evaluate and measure progress in language model reasoning!
Check out the full thread below!
Vishaal Udandarao@vishaal_urao
🚀New Paper! arxiv.org/abs/2504.07086 Everyone’s celebrating rapid progress in math reasoning with RL/SFT. But how real is this progress? We re-evaluated recently released popular reasoning models—and found reported gains often vanish under rigorous testing!! 👀 🧵👇
English
Bethge Lab รีทวีตแล้ว
🧠 Keeping LLMs factually up to date is a common motivation for knowledge editing.
But what would it actually take to support this in practice at the scale and speed the real world demands?
We explore this question and really push the limits of lifelong knowledge editing.
👇
English
Bethge Lab รีทวีตแล้ว
For our "Automated Assessment of Teaching Quality" project, we are looking for two PhD students: one in educational/cognitive sciences or a related field (uni-tuebingen.de/fakultaeten/wi…) and one in machine learning (uni-tuebingen.de/en/faculties/f…). Please apply and reach me out for details!
English
Bethge Lab รีทวีตแล้ว
AI can generate correct-seeming hypotheses (and papers!). Brandolini's law states BS is harder to refute than generate. Can LMs falsify incorrect solutions? o3-mini (high) scores just 9% on our new benchmark REFUTE. Verification is not necessarily easier than generation 🧵
English