Suivi des pages liées
Entrer un nom de page pour voir les modifications faites récemment sur des pages liées vers ou depuis cette page (pour voir les membres d’une catégorie, entrez Catégorie:Nom de catégorie). Les modifications des pages de votre liste de suivi sont en gras.
Légende des abréviations :
- N
- Cette modification a créé une nouvelle page (voir aussi la liste des nouvelles pages)
- m
- Cette modification est mineure.
- b
- Cette modification a été effectuée par un robot.
- (±123)
- La taille de la page a changé de ce nombre d’octets.
25 août 2025
| N 11:51 | Controlled NOT Gate diffhist +838 Arianne discussion contributions (Page créée avec « == en construction == == Définition == '''Porte logique quantique''' qui gère e flux d’informations entre deux '''qubits'''. Dans cette porte quantique, un qubit peut contrôler ce qui arrive à un autre qubit. == Français == ''' porte C-NOT''' ''' porte CNOT''' == Anglais == ''' C-NOT gate''' ''' CNOT gate''' ''' controlled NOT gate''' ''' controlled-NOT gate''' == Sources == [https://www.futura-sciences.com/sciences/ac... ») | ||||
| N 08:31 | Ovis diffhist +829 Pitpitt discussion contributions (Page créée avec « ==en construction== == Définition == XXXXXXXXX == Français == '''Ovis 2,5''' == Anglais == '''Ovis 2,5''' An advanced multimodal large language model designed to process images at their native resolutions while incorporating reasoning capabilities. The model addresses two key limitations in current vision-language systems: the degradation caused by fixed-resolution image processing and the lack of reflective reasoning beyond simple chain-of-thought approac... ») | ||||
| 08:29 | Journal des renommages Pitpitt discussion contributions a déplacé la page Nemotron Nano vers Nvidia Nemotron Nano | ||||
| N 08:28 | Nemotron Nano diffhist +672 Pitpitt discussion contributions (Page créée avec « ==en construction== == Définition == XXXXXXXXX == Français == ''' Nemotron Nano 2 ''' == Anglais == '''Nemotron Nano 2 ''' A hybrid Mamba-Transformer language model that combines high accuracy with significantly improved inference speed for reasoning tasks. The model achieves comparable or better performance than existing models while delivering up to 6× higher throughput for generation-heavy scenarios. This work demonstrates how architectural innovati... ») | ||||
| N 08:27 | ComoRAG diffhist +813 Pitpitt discussion contributions (Page créée avec « ==en construction== == Définition == XXXXXXXXX == Français == ''' ComoRAG''' == Anglais == '''ComoRAG''' ComoRAG, an iterative retrieval-based approach, enhances long-context narrative comprehension by dynamically updating memory and generating probing queries, outperforming traditional RAG methods. ComoRAG presents a cognitively-inspired approach to long narrative comprehension that addresses the stateless limitations of conventional RAG methods. By impl... ») | ||||
|
|
N 08:25 | Meta CLIP 2 modifications historique +1 014 [Pitpitt (2×)] | |||
|
|
08:25 (actu | diff) +1 Pitpitt discussion contributions | ||||
| N |
|
08:24 (actu | diff) +1 013 Pitpitt discussion contributions (Page créée avec « ==en construction== == Définition == XXXXXXXXX == Français == ''' Meta CLIP 2''' == Anglais == ''' Meta CLIP 2''' The first recipe for training CLIP models from scratch on worldwide web-scale image-text pairs spanning 300+ languages. The work addresses the challenge of scaling CLIP beyond English-only data while avoiding the "curse of multilinguality" - where multilingual models perform worse on English tasks than their English-only counterparts. Meta... ») | |||
20 août 2025
|
|
N 10:14 | Thyme 2 modifications historique +729 [Pitpitt (2×)] | |||
|
|
10:14 (actu | diff) +1 Pitpitt discussion contributions | ||||
| N |
|
10:14 (actu | diff) +728 Pitpitt discussion contributions (Page créée avec « ==en construction== == Définition == XXXXXXXXX == Français == ''' Thyme''' == Anglais == '''Thyme''' '''Think Beyond Images''' A novel approach that enables multimodal large language models to autonomously generate and execute code for diverse image processing operations and mathematical computations. Unlike existing "think with images" methods that are limited to simple cropping or image generation, Thyme provides rich functionality including rotation, c... ») | |||
19 août 2025
| 17:31 | Best-of-N Strategy diffhist +36 Arianne discussion contributions | ||||
| N 16:48 | Counterfactual Quantum Computation diffhist +343 Arianne discussion contributions (Page créée avec « == en construction == == Définition == xxxxx == Français == ''' XXXXXX''' == Anglais == ''' Counterfactual Quantum Computation''' == Sources == [https://arxiv.org/pdf/quant-ph/0607101v1 Source : arxiv] [https://en.wikipedia.org/wiki/Counterfactual_quantum_computation Source : Wikipedia] Catégorie:vocabulary Catégorie:Q2 ») | ||||
18 août 2025
| N 09:37 | DINO diffhist +1 349 Pitpitt discussion contributions (Page créée avec « ==en construction== == Définition == XXXXXXXXX == Français == ''' DINO v3 ''' == Anglais == '''DINO v3''' A self-supervised model trained without the need for manual data annotations. The method leverages simple yet effective strategies to scale both dataset and model size, achieving state-of-the-art performance across a broad range of vision tasks without requiring fine-tuning. The paper presents a versatile vision foundation model that significantly outp... ») | ||||
| N 09:36 | Mol-R1 diffhist +1 101 Pitpitt discussion contributions (Page créée avec « ==en construction== == Définition == XXXXXXXXX == Français == ''' Mol-R1''' == Anglais == '''Mol-R1''' Mol-R1 framework enhances molecule discovery by improving reasoning performance and explainability through PRID and MoIA strategies. A framework that enhances the reasoning capabilities of large language models for molecule discovery. The work addresses the challenge of generating molecular structures from text descriptions while providing clear, step-by... ») | ||||
15 août 2025
| N 09:24 | R-Zero diffhist +1 013 Pitpitt discussion contributions (Page créée avec « ==en construction== == Définition == XXXXXXXXX == Français == ''' R-Zero''' == Anglais == '''R-Zero''' A framework that enables large language models to improve their reasoning abilities without requiring any human-labeled training data. The method creates a self-evolving system where two AI models work together - one generates challenging questions while the other learns to solve them, creating an autonomous learning loop that starts from scratch. R-Z... ») | ||||
| N 09:23 | GLM diffhist +617 Pitpitt discussion contributions (Page créée avec « ==en construction== == Définition == XXXXXXXXX == Français == ''' XXXXXXXXX ''' == Anglais == '''GLM-4.5''' '''GLM''' GLM-4.5 represents a significant advancement in creating unified AI models that excel across multiple domains. By combining efficient MoE architecture, multi-stage training, and expert model iteration, the paper demonstrates that a single model can achieve strong performance in agentic tasks, reasoning, and coding without requiring the mas... ») | ||||
| N 09:20 | Omni-Effects diffhist +531 Pitpitt discussion contributions (Page créée avec « ==en construction== == Définition == XXXXXXXXX == Français == ''' XXXXXXXXX ''' == Anglais == '''Omni-Effects'''à A unified framework for generating customized visual effects (VFX) in videos. Unlike existing methods that require separate models for each effect, this approach can generate multiple visual effects simultaneously while providing precise spatial control over where each effect appears in the video. == Source == [https://huggingface.co/papers/... ») | ||||
| N 09:19 | WebWatcher diffhist +528 Pitpitt discussion contributions (Page créée avec « ==en construction== == Définition == XXXXXXXXX == Français == ''' XXXXXXXXX ''' == Anglais == '''WebWatcher''' A multimodal AI agent designed for deep research tasks that handles both visual and textual understanding. While existing web agents excel at text-based research, they struggle with real-world scenarios that involve visual information like scientific diagrams, charts, or visually rich web interfaces. == Source == [https://huggingface.co/papers/2... ») | ||||
12 août 2025
|
|
N 09:55 | MetaCLIP 3 modifications historique +999 [Pitpitt (3×)] | |||
|
|
09:55 (actu | diff) +93 Pitpitt discussion contributions | ||||
|
|
09:52 (actu | diff) +1 Pitpitt discussion contributions | ||||
| N |
|
09:52 (actu | diff) +905 Pitpitt discussion contributions (Page créée avec « ==en construction== == Définition == XXXXXXXXX == Français == ''' MetaCLIP''' == Anglais == '''MetaCLIP''' The first recipe for training CLIP models from scratch on worldwide web-scale image-text pairs spanning 300+ languages. The work addresses the challenge of scaling CLIP beyond English-only data while avoiding the "curse of multilinguality" - where multilingual models perform worse on English tasks than their English-only counterparts. The paper demonst... ») | |||
| N 09:50 | LongVie diffhist +672 Pitpitt discussion contributions (Page créée avec « ==en construction== == Définition == XXXXXXXXX == Français == ''' LongVie''' == Anglais == '''LongVie''' A framework for generating controllable long videos lasting up to one minute. The method addresses key challenges in extending video generation beyond short clips, specifically temporal inconsistency and visual degradation that occur when generating longer sequences. The paper proposes a multi-modal control approach that combines dense and sparse guidan... ») | ||||
| N 09:49 | PixNerd diffhist +689 Pitpitt discussion contributions (Page créée avec « ==en construction== == Définition == XXXXXXXXX == Français == ''' PixNerd''' == Anglais == '''PixNerd''' A novel approach to image generation that operates directly in pixel space rather than compressed latent representations. The method addresses limitations of current diffusion models that rely on variational autoencoders (VAEs), which can introduce artifacts and require complex two-stage training. By combining diffusion transformers with neural field re... ») | ||||
5 août 2025
| 15:30 | Mesure de distance sur des segments textuels diffhist −15 Patrickdrouin discussion contributions | ||||
4 août 2025
| N 09:47 | MUR diffhist +506 Pitpitt discussion contributions (Page créée avec « ==en construction== == Définition == XXXXXXXXX == Français == ''' MUR ''' == Anglais == '''MUR ''' MUR, a training-free approach that makes test-time scaling for language model reasoning both more efficient and more effective. By using momentum-based uncertainty tracking to identify critical reasoning steps, the method achieves substantial computational savings while improving accuracy. == Source == [https://huggingface.co/papers/2507.14958 Source : h... ») | ||||
| N 09:45 | Goedel-Prover diffhist +612 Pitpitt discussion contributions (Page créée avec « ==en construction== == Définition == XXXXXXXXX == Français == ''' Goedel-Prover''' == Anglais == '''Goedel-Prover''' An open-source model for automated theorem proving. To overcome the scarcity of formal math data, they first trained models to convert 1.64 million natural language math problems into the formal language Lean 4. They then used an "expert iteration" process, where successive versions of the prover generate new proofs that are verified and... ») | ||||
| N 09:44 | SPIRAL diffhist +352 Pitpitt discussion contributions (Page créée avec « ==en construction== == Définition == XXXXXXXXX == Français == ''' SPIRAL''' == Anglais == '''SPIRAL''' A framework where language models develop reasoning skills by playing zero-sum games against themselves, removing the need for human-supervised datA == Source == [https://arxiv.org/pdf/2506.24119 Source : arxiv] Catégorie:vocabulary ») | ||||
| N 09:41 | Voice cloning diffhist +450 Pitpitt discussion contributions (Page créée avec « ==en construction== == Définition == XXXXXXXXX == Français == ''' XXXXXXXXX ''' == Anglais == '''Voice cloning''' À The method of using AI models to create a digital copy of a person’s voice and then potentially generating new speech samples in that voice. == Source == [https://www.newscientist.com/article/2445475-how-to-avoid-being-fooled-by-ai-generated-misinformation/#DeepDive-1 Source : newscientist] Catégorie:vocabulary ») | ||||
