[GUEST ACCESS MODE: Data is scrambled or limited to provide examples. Make requests using your API key to unlock full data. Check https://lunarcrush.ai/auth for authentication information.]
@ManuelFaysse Manuel Faysse
Manuel Faysse posts on X about wrt the most. They currently have XXXXX followers and XX posts still getting attention that total XXX engagements in the last XX hours.
Engagements: XXX #
- X Week XXXXXX +8,542%
- X Month XXXXXX +8,569%
- X Months XXXXXX -XX%
- X Year XXXXXXX +50,181%
Mentions: X #
Followers: XXXXX #
- X Week XXXXX +6.30%
- X Month XXXXX +11%
- X Months XXXXX +30%
- X Year XXXXX +319%
CreatorRank: XXXXXXXXX #
Social Influence #
Social topic influence wrt
Top Social Posts #
Top posts by engagements in the last XX hours
"@gigant_theo @antonio_loison @pteiletche Normally VLMs really attenuated the modality gap problem. Did you check out the latest Jina embeddings Or cohere X Everything is projected in the same latent space which is the one of the text model. The scoring func still might be a bit biased but VLMs helped a lot"
@ManuelFaysse on X 2025-07-17 09:36:22 UTC 1943 followers, XXX engagements
"Introducing ColQwen-Omni a 3B omnimodal retriever that extends the ColPali concept of multimodal retrieval with late interaction to audio chunks and short videos with no performance degradation on visual document retrieval wrt our best models (1/N)" @ManuelFaysse on X 2025-07-17 08:14:28 UTC 1947 followers, 54.1K engagements
"This is a v0 and as we specifically train to improve audio retrieval performance will improve. Eventually we aim to investigate cross-modality performance transfer and create a great all-around retriever with @antonio_loison @pteiletche + Quentin. (5/N)" @ManuelFaysse on X 2025-07-17 08:14:30 UTC 1948 followers, 1202 engagements
"Thanks to the Qwen team for a great backbone model adapting it for retrieval was super quick to do but it's hard to find the time :/ For those interested @xueguang_ma also trained a single-vector version to check out (6/N)" @ManuelFaysse on X 2025-07-17 08:14:31 UTC 1947 followers, XXX engagements
"The model is trained solely on visual document retrieval. Based on the Qwen-Omni backbone it is able to generalize its newly learned document embedding capabilities to embed audios and short videos without ever having seen those during contrastive training (2/N)" @ManuelFaysse on X 2025-07-17 08:14:29 UTC 1948 followers, 1626 engagements