
Daniel Cer
@daniel_m_cer
Followers
559
Following
3K
Media
8
Statuses
178
Research Scientist at @GoogleAI, @googIeresearch.
California, USA
Joined March 2012
RT @tonywu_71: Similarity maps also works for the Hf-native ColQwen2 model! š¤ I have created a cookbook to quickly try this out:. https://tā¦.
github.com
Recipes for learning, fine-tuning, and adapting ColPali to your multimodal RAG use cases. šØš»āš³ - tonywu71/colpali-cookbooks
0
5
0
RT @sigridjin_eth: @daniel_m_cer Here is the Python Implementation.
github.com
The Python Implementation of CRISP: Clustering Multi-Vector Representations for Denoising and Pruning - sigridjineth/crisp-py
0
2
0
RT @MaxSonate: Our Gemini model just won a gold medal at the IMO 2025. Itās a massive milestone for AI, and Iām so proud to have played a pā¦.
0
8
0
RT @lateinteraction: š¢ If youāre at #SIGIR2025 this week, make sure to be at Luca Scheererās paper talk:. āWARP: An Efficient Engine for Muā¦.
0
20
0
RT @tomaarsen: ā¼ļøSentence Transformers v5.0 is out! The biggest update yet introduces Sparse Embedding models, encode methods improvements,ā¦.
0
66
0
RT @beirmug: We had a successful participation of over 45+ teams & 150+ runs last year in TREC RAG 2024! š„š„. We are back with @TREC_RAG 202ā¦.
docs.google.com
We are having some systems issues, so in the meantime, we are taking registrations through this form. Please use a direct email address and not a mailing list or group alias as the contact email.
0
3
0
RT @harveyiyun: LLMs excel at finding surprising āneedlesā in very long documents, but can they detect when information is conspicuously miā¦.
0
32
0
RT @oscrhong: Interesting tidbit from prof @chrmanning: The first mention of āLarge Language Modelā comes from a 1998 NLP workshop Taiwan!ā¦.
0
7
0
RT @GoogleResearch: Neural embedding models have become a cornerstone of modern information retrieval. Today we introduce MUVERA, a state-oā¦.
0
79
0
RT @_reachsumit: Video-ColBERT: Contextualized Late Interaction for Text-to-Video Retrieval. Introduces a bi-encoder approach that performsā¦.
arxiv.org
In this work, we tackle the problem of text-to-video retrieval (T2VR). Inspired by the success of late interaction techniques in text-document, text-image, and text-video retrieval, our approach,...
0
3
0
RT @meetdavidwan: Excited to share our new work, CLaMR! š. We tackle multimodal content retrieval by jointly considering video, speech, OCRā¦.
0
65
0
RT @bclavie: Multimodal RAG: Just use ColPali/DSE then pass your screenshots to the LLM. This is the dream, but how well do LLMs read textā¦.
0
82
0
RT @raphaelsrty: I'm thrilled to announce the release of FastPlaid ! šš. FastPlaid is a high-performance engine for multi-vector search, buā¦.
0
38
0
RT @tonywu_71: š ColQwen2 just dropped in Transformers! š¤. Say goodbye to brittle OCR pipelines ā now you can retrieve documents directly iā¦.
0
94
0
RT @davidbau: Dear MAGA friends,. I have been worrying about STEM in the US a lot, because right now the Senate is writing new laws that cuā¦.
0
72
0
RT @_reachsumit: R3-RAG: Learning Step-by-Step Reasoning and Retrieval for LLMs via Reinforcement Learning. Introduces RL to teach LLMs adaā¦.
github.com
Contribute to Yuan-Li-FNLP/R3-RAG development by creating an account on GitHub.
0
2
0
RT @ManuelFaysse: šØ Context matters for effective retrievalābut most embedding models cannot leverage crucial information outside of the paā¦.
0
30
0
RT @younggyoseo: Excited to present FastTD3: a simple, fast, and capable off-policy RL algorithm for humanoid control -- with an open-sourcā¦.
0
115
0
RT @beirmug: Did you know that fine-tuning retrievers & re-rankers on large but unclean training datasets can harm their performance? š”. Inā¦.
0
19
0
RT @lateinteraction: Google folks continues to do awesome late interaction work. Compared to vanilla ColBERT, a version of this new āCRISPā¦.
0
14
0