domain-GenAI

Domain: GenAI (105)

video-gen (51)

• its-been-a-busy-week-for-world-models-and-their-apis — It’s been a busy week for World Models and their APIs! https://t.co/NDHLgQUV6O…
• new-nvidia-paper-we-introduce-motive-a-motion-centric — New NVIDIA Paper We introduce Motive, a motion-centric, gradient-based data at…
• physrvg-physics-aware-unified-reinforcement-learning-for — PhysRVG Physics-Aware Unified Reinforcement Learning for Video Generative Model…
• what-if-an-ai-could-generate-a-5-minute-video-from-a-single — what-if-an-ai-could-generate-a-5-minute-video-from-a-single
• world-models-dont-fail-because-they-cant-predict-the-future — world-models-dont-fail-because-they-cant-predict-the-future
• stream-diffvsr-real-time-diffusion-based-video-super — stream-diffvsr-real-time-diffusion-based-video-super
• stream-diffvsr-a-low-latency-streamable-video-super — stream-diffvsr-a-low-latency-streamable-video-super
• tedla-et-al-generating-the-past-present-and-future-from-a — tedla-et-al-generating-the-past-present-and-future-from-a
• egox-generate-immersive-first-person-video-from-any-third-pe — egox-generate-immersive-first-person-video-from-any-third-pe
• zeng-et-al-neuralremaster-phase-preserving-diffusion-for-str — zeng-et-al-neuralremaster-phase-preserving-diffusion-for-str
• bullettime-decoupled-control-of-time-and-camera-pose-for-vid — bullettime-decoupled-control-of-time-and-camera-pose-for-vid
• introducing-relic-world-model-real-time-interactive-video-ge — introducing-relic-world-model-real-time-interactive-video-ge
• video4spatial-towards-visuospatial-intelligence-with-context — video4spatial-towards-visuospatial-intelligence-with-context
• excited-to-share-our-neurips2025-paper-physctrl-generative-p — excited-to-share-our-neurips2025-paper-physctrl-generative-p
• meet-gaia-3-wayves-most-advanced-generative-world-model-yet- — meet-gaia-3-wayves-most-advanced-generative-world-model-yet-
• pan-a-world-model-for-general-interactable-and-long-horizon- — pan-a-world-model-for-general-interactable-and-long-horizon-
• nano-banana-makeugc-veo3-ai-content-factory-this-agent-pumps — nano-banana-makeugc-veo3-ai-content-factory-this-agent-pumps
• motionstream-real-time-interactive-video-generation-with-mou — motionstream-real-time-interactive-video-generation-with-mou
• generative-view-stitching-we-worked-with-to-tame-the-short-s — generative-view-stitching-we-worked-with-to-tame-the-short-s
• zero-shot-video-reasoning-chain-of-frames-isnt-just-for — zero-shot-video-reasoning-chain-of-frames-isnt-just-for
• code2video-a-code-centric-paradigm-for-educational-video — code2video-a-code-centric-paradigm-for-educational-video
• want-to-build-𝟑𝐃-𝐅𝐨𝐮𝐧𝐝𝐚𝐭𝐢𝐨𝐧-𝐌𝐨𝐝𝐞𝐥𝐬 — want-to-build-𝟑𝐃-𝐅𝐨𝐮𝐧𝐝𝐚𝐭𝐢𝐨𝐧-𝐌𝐨𝐝𝐞𝐥𝐬
• 4dnex-feed-forward-4d-generative-modeling-made-easy — 4dnex-feed-forward-4d-generative-modeling-made-easy
• 1957439501644562640 — CARTOON HERO 1.0 - AI Animation System
• 1957182570309013714 — Tencent Hunyuan Open Source Alternative to Genie 3
• i-asked-veo-3-to-create-video-game-scenes-to-famous-places-results-are-wild-10 — i-asked-veo-3-to-create-video-game-scenes-to-famous-places-results-are-wild-10
• epona-autoregressive-diffusion-world-model-for-autonomous-driving — epona-autoregressive-diffusion-world-model-for-autonomous-driving
• immersegen-agent-guided-immersive-world-generation-with-alpha-textured-proxies-h — immersegen-agent-guided-immersive-world-generation-with-alpha-textured-proxies-h
• introducing-unirelight-a-general-purpose-relighting-framework-powered-by-video — introducing-unirelight-a-general-purpose-relighting-framework-powered-by-video
• -excited-to-introduce-simworld-an-embodied-simulator-for-infinite-photorealistic — -excited-to-introduce-simworld-an-embodied-simulator-for-infinite-photorealistic
• the-context-size-of-video-world-models-is-only-a-few-frames-like-a-human-with-se — the-context-size-of-video-world-models-is-only-a-few-frames-like-a-human-with-se
• preprint-of-today-sun-et-al-unigeo-taming-video-diffusion-for-unified-consistent — preprint-of-today-sun-et-al-unigeo-taming-video-diffusion-for-unified-consistent
• preprint-of-today-wang-et-al-ati-any-trajectory-instruction-for-controllable-vid — preprint-of-today-wang-et-al-ati-any-trajectory-instruction-for-controllable-vid
• frame-in-n-out-unbounded-controllable-image-to-video-generation-httpstcopoiu2jr6 — frame-in-n-out-unbounded-controllable-image-to-video-generation-httpstcopoiu2jr6
• vid2world-crafting-video-diffusion-models-to-interactive-world-models-httpstcoad — vid2world-crafting-video-diffusion-models-to-interactive-world-models-httpstcoad
• nvidia-just-dropped-gen3c-3d-informed-world-consistent-video-generation-with-pre — nvidia-just-dropped-gen3c-3d-informed-world-consistent-video-generation-with-pre
• excited-to-introduce-gen3c-cvpr2025-a-generative-video-model-with-an-explicit-3d — excited-to-introduce-gen3c-cvpr2025-a-generative-video-model-with-an-explicit-3d
• wow-the-new-skyreels-video-model-allows-for-really-precise-editing-via-flowedit — wow-the-new-skyreels-video-model-allows-for-really-precise-editing-via-flowedit
• introducing-diffusionrenderer-a-neural-rendering-engine-powered-by-video — introducing-diffusionrenderer-a-neural-rendering-engine-powered-by-video
• dreamdrive-generative-4d-scene-modeling-from-street-view-images-pointscoder-boyi — dreamdrive-generative-4d-scene-modeling-from-street-view-images-pointscoder-boyi
• big-odysseyml-news-were-unveiling-explorer-a-generative-world-model-explorer-tra — big-odysseyml-news-were-unveiling-explorer-a-generative-world-model-explorer-tra
• 3dtrajmaster-mastering-3d-trajectory-for-multi-entity-motion-in-video-generation — 3dtrajmaster-mastering-3d-trajectory-for-multi-entity-motion-in-video-generation
• after-reconx-further-explore-the-potential-of-video-diffusion-and-propose-dimens — after-reconx-further-explore-the-potential-of-video-diffusion-and-propose-dimens
• google-presents-vidpanos-generative-panoramic-videos-from-casual-panning-videos — google-presents-vidpanos-generative-panoramic-videos-from-casual-panning-videos
• diffusion-models-continue-to-get-stronger-reconx-reconstruct-any-scene-from — diffusion-models-continue-to-get-stronger-reconx-reconstruct-any-scene-from
• animate3d-animating-any-3d-model-with-multi-view-video-diffusion-recent — animate3d-animating-any-3d-model-with-multi-view-video-diffusion-recent
• can-generative-models-synthesize-policy-networks-for-agents-like-text-to-image-g — can-generative-models-synthesize-policy-networks-for-agents-like-text-to-image-g
• can-generative-models-synthesize-policy-networks-for-agents-like-text-to-image — can-generative-models-synthesize-policy-networks-for-agents-like-text-to-image
• heading-to-in-please-dm-me-if-youd-like-to-have-a-chat-espec — heading-to-in-please-dm-me-if-youd-like-to-have-a-chat-espec
• fine-grained-controllable-video-generation-via-object-appearance-and-context-pap — fine-grained-controllable-video-generation-via-object-appearance-and-context-pap
• physics-aware-unified-reinforcement-learning-for-video-generative-models — Physics-Aware Unified Reinforcement Learning for Video Generative Models

diffusion (14)

• just-found-the-dllm-library-to-create-diffusion-language-mod — just-found-the-dllm-library-to-create-diffusion-language-mod
• added-confidence-aware-parallel-decoding-to-my-tiny-text-dif — added-confidence-aware-parallel-decoding-to-my-tiny-text-dif
• tired-to-go-back-to-the-original-papers-again-and-again-our — tired-to-go-back-to-the-original-papers-again-and-again-our
• ominicontrol-has-been-presented-as-a-highlight-at-iccv-2025 — ominicontrol-has-been-presented-as-a-highlight-at-iccv-2025
• 1955901155726516652 — Google Nano-Banana Image Model Results
• flow-matching-fm-is-one-of-the-hottest-ideas-in-generative-ai---and-its — flow-matching-fm-is-one-of-the-hottest-ideas-in-generative-ai---and-its
• diffusion-모델은-aaa-게임-개발의-새로운-선구자입니다---diffusion-기반-생성형-ai-모델이-차세대-aaa-게 — diffusion-모델은-aaa-게임-개발의-새로운-선구자입니다---diffusion-기반-생성형-ai-모델이-차세대-aaa-게
• normally-changing-robot-policy-behavior-means-changing-its-weights-or-relying — normally-changing-robot-policy-behavior-means-changing-its-weights-or-relying
• want-to-generate-realistic-handobject-manipulations-for-unseen-objects-check-out — want-to-generate-realistic-handobject-manipulations-for-unseen-objects-check-out
• google-presents-lightlab-controlling-light-sources-in — google-presents-lightlab-controlling-light-sources-in
• how-much-data-do-you-need-to-train-an-ai-model-to-generate-realistic-photos — how-much-data-do-you-need-to-train-an-ai-model-to-generate-realistic-photos
• cosmos-is-a-developer-first-platform-designed-to-help-physical-ai-builders-accel — cosmos-is-a-developer-first-platform-designed-to-help-physical-ai-builders-accel
• this-is-a-single-uncut-video-showing-a-robot-learning-several-tasks-instantly-af — this-is-a-single-uncut-video-showing-a-robot-learning-several-tasks-instantly-af
• deep-learning-for-computer-vision-dl4cv-learn-about-modern-methods-for-computer — deep-learning-for-computer-vision-dl4cv-learn-about-modern-methods-for-computer

image-gen (13)

• ubisoft-la-forge-open-sourced-its-pbr-material-model-chord-t — ubisoft-la-forge-open-sourced-its-pbr-material-model-chord-t
• 1960464912758493410 — Nano Banana + Runway Act 2 for Creative Flexibility
• -what-if-gaud-had-midjourney-studiotimfu-s-living-sketches-explore-architecture — -what-if-gaud-had-midjourney-studiotimfu-s-living-sketches-explore-architecture
• chinas-tencent-just-dropped-hunyuancustom-this-ai-turns-any — chinas-tencent-just-dropped-hunyuancustom-this-ai-turns-any
• this-object-removal-workflow-kinda-slaps-thoughts-link-to — this-object-removal-workflow-kinda-slaps-thoughts-link-to
• a-deep-mathematical-exploration-of-diffusion-models-for-modern-ai-image-generati — a-deep-mathematical-exploration-of-diffusion-models-for-modern-ai-image-generati
• nvidia-presents-add-it-training-free-object-insertion-in-images-with-pretrained — nvidia-presents-add-it-training-free-object-insertion-in-images-with-pretrained
• generating-images-with-4-bit-flux-schnell-on-my-m1-max-laptop-is-pretty-awesome-1 — generating-images-with-4-bit-flux-schnell-on-my-m1-max-laptop-is-pretty-awesome-1
• generating-images-with-4-bit-flux-schnell-on-my-m1-max-laptop-is-pretty-awesome — generating-images-with-4-bit-flux-schnell-on-my-m1-max-laptop-is-pretty-awesome
• mlx-flux-with-mlxdiffusionkit-amazingly-it-works-on-mac-mini16gb-even-though — mlx-flux-with-mlxdiffusionkit-amazingly-it-works-on-mac-mini16gb-even-though
• mlx-flux-with-mlxdiffusionkit-amazingly-it-works-on-mac-mini16gb-even-though-the — mlx-flux-with-mlxdiffusionkit-amazingly-it-works-on-mac-mini16gb-even-though-the
• physical-light-real-time-diffusion-exploration-a-thread-1n — physical-light-real-time-diffusion-exploration-a-thread-1n
• excited-to-share-our-work-on-neural-assets-a-new-method-for-enabling-3d-asset — excited-to-share-our-work-on-neural-assets-a-new-method-for-enabling-3d-asset

text-to-X (10)

• unsloth-has-a-great-guide-on-fine-tuning-llms-selecting-the- — unsloth-has-a-great-guide-on-fine-tuning-llms-selecting-the-
• 와-이거-정말-좋은데-안-그래도-rag-구축할-때마다-왜-굳이-벡터로-돌아가는-llm에게-컨텍 — 와-이거-정말-좋은데-안-그래도-rag-구축할-때마다-왜-굳이-벡터로-돌아가는-llm에게-컨텍
• openai-devday-키노트-지극히-개인적인-요약-1-apps-sdk-gpts-action-에다가 — openai-devday-키노트-지극히-개인적인-요약-1-apps-sdk-gpts-action-에다가
• this-is-the-answer-to-how-humans-and-machines-collaborate-in-the-generative-ai-e — this-is-the-answer-to-how-humans-and-machines-collaborate-in-the-generative-ai-e
• -we — -we
• announcing-our-13m-funding-round-to-build-the-next-generation-of-ai — announcing-our-13m-funding-round-to-build-the-next-generation-of-ai
• for-a-collection-of-advanced-retrieval-augmented-generation-rag-techniques-this-1 — for-a-collection-of-advanced-retrieval-augmented-generation-rag-techniques-this-1
• clip-is-the-default-choice-for-most-multimodal-llm-research-but-we-know-clip-is-1 — clip-is-the-default-choice-for-most-multimodal-llm-research-but-we-know-clip-is-1
• multimodal-llms-are-just-superb-to-play-with-hacking-around-with-qwen2-vl-and-it — multimodal-llms-are-just-superb-to-play-with-hacking-around-with-qwen2-vl-and-it
• pre-releasing-dynamic-rendering-in-gradio-the-idea-is-that-n — pre-releasing-dynamic-rendering-in-gradio-the-idea-is-that-n

3D-gen (6)

• tinker-diffusions-gift-to-3d-multi-view-consistent-editing-from-sparse-input — tinker-diffusions-gift-to-3d-multi-view-consistent-editing-from-sparse-input
• monocular-dynamic-reconstruction-is-hard-single-video-input-gt-a-lot-of-3d — monocular-dynamic-reconstruction-is-hard-single-video-input-gt-a-lot-of-3d
• interested-in-3d-vision-3d-scene-understanding-and-3d-generation-we-have-open-ph — interested-in-3d-vision-3d-scene-understanding-and-3d-generation-we-have-open-ph
• cvpr-2025-paper-alert-paper-title-magicarticulate-make-your-3d-models-articulati — cvpr-2025-paper-alert-paper-title-magicarticulate-make-your-3d-models-articulati
• real2code-translating-real-world-articulated-objects-to-sim — real2code-translating-real-world-articulated-objects-to-sim
• using-gradio-to-generate-3d-objects-for-your-vision-pro — using-gradio-to-generate-3d-objects-for-your-vision-pro

visionos (5)

• how-can-we-create-interactive-physical-digital-twins-from-videos — how-can-we-create-interactive-physical-digital-twins-from-videos
• want-to-generate-bimanual-hand-interactions-given-an-articulated-object — want-to-generate-bimanual-hand-interactions-given-an-articulated-object
• 13-zerocomp-is-being-presented-as-an-oral-today-at-wacv2025 — 13-zerocomp-is-being-presented-as-an-oral-today-at-wacv2025
• step-by-step-diffusion-an-elementary-tutorial — step-by-step-diffusion-an-elementary-tutorial
• 𝗗𝗮𝘁𝗮-𝗣𝗶𝗽𝗲𝗹𝗶𝗻𝗲𝘀-𝗶𝗻-𝗠𝗮𝗰𝗵𝗶𝗻𝗲-𝗟𝗲𝗮𝗿𝗻𝗶𝗻𝗴-𝗦𝘆𝘀𝘁𝗲𝗺𝘀-can-become-complex-and-for-a-good — 𝗗𝗮𝘁𝗮-𝗣𝗶𝗽𝗲𝗹𝗶𝗻𝗲𝘀-𝗶𝗻-𝗠𝗮𝗰𝗵𝗶𝗻𝗲-𝗟𝗲𝗮𝗿𝗻𝗶𝗻𝗴-𝗦𝘆𝘀𝘁𝗲𝗺𝘀-can-become-complex-and-for-a-good

3d-gen (3)

• 1965495039217504464 — AI-Powered Automatic PBR Texturing for Any Mesh
• 1963580239541363098 — 3D Foundation Models Hiring - Munich & London
• 1959608339337281586 — Future of Architecture with AI - Image to 3D

editing (1)

• 1948878604928254257 — Aleph Instantaneous Inpainting - Remove Reflections

web-graphics (1)

• blenderfusion-3d-grounded-visual-editing-and-generative-compositing — blenderfusion-3d-grounded-visual-editing-and-generative-compositing

reconstruction (1)

• can-generative-video-models-help-pose-estimation-yes-we-find-that-generative-vid — can-generative-video-models-help-pose-estimation-yes-we-find-that-generative-vid