enriched-codex

🤖 Codex Enriched (246)

gpt-5.3-codex로 분류된 북마크 목록

3D-Vision (72)

• 1961148220358397984 — LSD-3D: Large-Scale 3D Driving Scene Generation with Geometry Grounding
• 1965013985927082013 — MILo - Mesh-In-the-Loop Gaussian Splatting for Mesh Extraction
• 1n-rotary-position-embeddings-rope-are-ubiquitous-across-tra — 1n-rotary-position-embeddings-rope-are-ubiquitous-across-tra
• 1n-rotary-position-embeddings-rope-are-ubiquitous-across-transformers-that — RayRoPE: Projective Ray Positional Encoding for Multi-view Attention
• 1n-will-this-be-the-bertgpt-moment-for-3d-vision — 1n-will-this-be-the-bertgpt-moment-for-3d-vision
• 2d-triangle-splatting-for-direct-differentiable-mesh-training-kaifeng-sheng — 2d-triangle-splatting-for-direct-differentiable-mesh-training-kaifeng-sheng
• 30fps720p-i-think-we-got-just-the-right-paper-for-you-give-us-a-week-for-arxiv — 30fps720p-i-think-we-got-just-the-right-paper-for-you-give-us-a-week-for-arxiv
• 315-pm-515-pm-exhibit-hall-i-767 — 315-pm-515-pm-exhibit-hall-i-767
• 360anything-geometry-free-lifting-of-images-and-videos-to-360 — 360Anything: Geometry-Free Lifting of Images and Videos to 360°
• 360anything-geometry-free-lifting-of-images-and-videos-to — 360Anything: Geometry-Free Lifting of Images and Videos to 360°
• 3d-gaussian-splatting-3dgs-is-becoming-the-new-standard-for-3d-reconstruction-ra — 3d-gaussian-splatting-3dgs-is-becoming-the-new-standard-for-3d-reconstruction-ra
• 3d-gaussian-splatting-for-large-scale-3d-surface-reconstruction-from-aerial-imag — 3d-gaussian-splatting-for-large-scale-3d-surface-reconstruction-from-aerial-imag
• 3d-gaussian-splatting-for-large-scale-3d-surface-reconstruction-from-aerial — 3d-gaussian-splatting-for-large-scale-3d-surface-reconstruction-from-aerial
• 3d-gaussian-splatting-is-awesome-but-the-3d-gaussian-primitives-often-have-diffi — 3d-gaussian-splatting-is-awesome-but-the-3d-gaussian-primitives-often-have-diffi
• 3d-gaussian-splatting-is-definitely-the-future-its-going-to-allow-us-to-efficien — 3d-gaussian-splatting-is-definitely-the-future-its-going-to-allow-us-to-efficien
• 3d-llm-injecting-the-3d-world-into-large-language-models — 3d-llm-injecting-the-3d-world-into-large-language-models
• 3d-object-tracking-without-training-data-in-our-machine-intelligence-pa — Towards generalizable and interpretable three-dimensional tracking with inverse neural rendering
• 3d-re-gen-3d-reconstruction-of-indoor-scenes-with-a-1 — 3D-RE-GEN: 3D Reconstruction of Indoor Scenes with a Generative Framework
• 3d-re-gen-3d-reconstruction-of-indoor-scenes-with-a — 3D-RE-GEN 3D Reconstruction of Indoor Scenes with a Generative Framework
• 3d-reconstruction-with-spatial-memory-discuss-we-present-spann3r-a-novel-approac — 3D Reconstruction with Spatial Memory
• 3d-reconstruction-with-spatial-memory-discuss-we-present-spann3r-a-novel — 3D Reconstruction with Spatial Memory
• 3d-vision-language-gaussian-splatting-qucheng-peng-benjamin-planche-zhongpai-gao — 3D Vision-Language Gaussian Splatting
• 3d-vision-language-gaussian-splatting-qucheng-peng-benjamin-planche-zhongpai — 3D Vision-Language Gaussian Splatting
• 3dgs-in-the-video-created-by-the-world-class-splat-scanner-christoph-schindelar — Christoph Schindelar 3DGS 스캔 영상
• 3dgs2-tr-scalable-second-order-trust-region-method-for-3d-ga — 3DGS^2-TR: Scalable Second-Order Trust-Region Method for 3D Gaussian Splatting
• 3dgs2-tr-scalable-second-order-trust-region-method-for-3d-gaussian-splatting — 3DGS²-TR: Scalable Second-Order Trust-Region Method for 3D Gaussian Splatting
• 4d3r-motion-aware-neural-reconstruction-and-rendering-of-dyn — 4D3R: Motion-Aware Neural Reconstruction and Rendering of Dynamic Scenes from Monocular Videos
• 4dgt-learning-a-4d-gaussian-transformer-using-real-world-monocular-videos-abstra — 4DGT: Learning a 4D Gaussian Transformer Using Real-World Monocular Videos
• 4dhands-can-robustly-recover-interactive-hand-meshes-and-the — 4DHands: Reconstructing Interactive Hands in 4D with Transformers
• a-great-example-of-using-vggt-inside-blender-please-check — 블렌더용 VGGT 통합 데모
• a-new-milestone-for-real-time-accurate-3d-spatial-computing — Fast-FoundationStereo: 실시간 스테레오 깊이 추정
• a-radiance-field-loss-for-fast-and-simple-emissive-surface — A Radiance Field Loss for Fast and Simple Emissive Surface Reconstruction
• a-scene-is-worth-a-thousand-features-feed-forward-camera — A Scene is Worth a Thousand Features: Feed-Forward Camera Localization from a Collection of Image Features
• a-survey-on-3d-gaussian-splatting-applications-segmentation-editing-and-gener — A Survey on 3D Gaussian Splatting Applications: Segmentation, Editing, and Generation
• about-halfway-through-building-out-the-slides-for-my — Gaussian Splatting 강의 슬라이드 작업
• ace-slam-scene-coordinate-regression-for-real-time-slam — ACE-SLAM: Scene Coordinate Regression for Neural Implicit Real-Time SLAM
• actionmesh — ActionMesh: Animated 3D Mesh Generation with Temporal 3D Diffusion
• activesplat-high-fidelity-scene-reconstruction-through-active-gaussian-splatting — ActiveSplat: High-Fidelity Scene Reconstruction through Active Gaussian Splatting
• adagar-a-unified-framework-for-dynamic-3d-scene — AdaGaR: Adaptive Gabor Representation for Dynamic Scene Reconstruction
• advances-in-feed-forward-3d-reconstruction-and-view-synthesis-a-survey-abstract — Advances in Feed-Forward 3D Reconstruction and View Synthesis: A Survey
• advances-in-radiance-field-for-dynamic-scene-from-neural — Advances in Radiance Field for Dynamic Scene: From Neural Field to Gaussian Field
• after-a-little-troubleshooting-i-got-glomap-to-produce-this-sparse-reconstructio — GLOMAP 희소 재구성 속도
• after-a-little-troubleshooting-i-got-glomap-to-produce-this-sparse — GLOMAP 70초 희소 재구성
• after-two-incredible-years-at-meta-its-time-for-a-new-adventure-its-been-a-journ — Meta 2년 3D 연구 여정 마무리
• aha-animating-human-avatars-in-diverse-scenes-with-gaussian — AHA! Animating Human Avatars in Diverse Scenes with Gaussian Splatting
• and-just-when-i-thought-having-to-repeatedly-say-vggt-was — TTT3R: 3D Reconstruction as Test-Time Training
• animate3d-animating-any-3d-model-with-multi-view-video-diffusion-recent-advances — Animate3D: Animating Any 3D Model with Multi-view Video Diffusion
• animated-3d-mesh-generation-with-temporal-3d-diffusion-httpstcowi22ijg3wr — ActionMesh: Animated 3D Mesh Generation with Temporal 3D Diffusion
• artdeco-towards-efficient-and-high-fidelity-on-the-fly-3d — ARTDECO: Towards Efficient and High-Fidelity On-the-Fly 3D Reconstruction with Structured Scene Representation
• bringing-foundation-models-to-depth-sensing-defm-is-trained- — DeFM
• bulletgen-improving-4d-reconstruction-with-bullet-time-generation-numair-khan — bulletgen-improving-4d-reconstruction-with-bullet-time-generation-numair-khan
• camera-splatting-for-continuous-view-optimization — camera-splatting-for-continuous-view-optimization
• cameras-as-relative-positional-encoding-tldr-study-absolute-raymap-and-relative — cameras-as-relative-positional-encoding-tldr-study-absolute-raymap-and-relative
• career-update-happy-to-share-that-ive-joined-meta-reality-labs-as-a-research-sci — career-update-happy-to-share-that-ive-joined-meta-reality-labs-as-a-research-sci
• check-out-our-cvpr2025-paper-with-all-5 — check-out-our-cvpr2025-paper-with-all-5
• check-out-the-efm3d-benchmark-and-try-to-beat-our-evl-models-performance — check-out-the-efm3d-benchmark-and-try-to-beat-our-evl-models-performance
• cl-splats-continual-learning-of-gaussian-splatting-with-local-optimization-tldr — cl-splats-continual-learning-of-gaussian-splatting-with-local-optimization-tldr
• cmon-guys-youre-all-sleeping-on-innovation-how-did-this — cmon-guys-youre-all-sleeping-on-innovation-how-did-this
• code-dropped-1 — code-dropped-1
• code-dropped-2 — code-dropped-2
• code-dropped-and-accepted-at-iclr-2025 — code-dropped-and-accepted-at-iclr-2025
• come-by-our-poster-this-afternoon-𝟭𝟳𝟭𝟱-𝘁𝗼-𝟭𝟴𝟰𝟱-𝗔𝗿𝗰𝗵-𝟰𝗔-𝗘 — come-by-our-poster-this-afternoon-𝟭𝟳𝟭𝟱-𝘁𝗼-𝟭𝟴𝟰𝟱-𝗔𝗿𝗰𝗵-𝟰𝗔-𝗘
• como-will-be-presented-as-an-𝐨𝐫𝐚𝐥-at-also-come-see-the-𝐥𝐢𝐯𝐞-𝐝𝐞𝐦𝐨-of-our-compact — como-will-be-presented-as-an-𝐨𝐫𝐚𝐥-at-also-come-see-the-𝐥𝐢𝐯𝐞-𝐝𝐞𝐦𝐨-of-our-compact
• como-will-be-presented-as-an-𝐨𝐫𝐚𝐥-at-eccv2024-also-come-see-the-𝐥𝐢𝐯𝐞-𝐝𝐞𝐦𝐨-of — como-will-be-presented-as-an-𝐨𝐫𝐚𝐥-at-eccv2024-also-come-see-the-𝐥𝐢𝐯𝐞-𝐝𝐞𝐦𝐨-of
• considering-embedding-knowledge-from-foundation-models-into-3d-gaussian — considering-embedding-knowledge-from-foundation-models-into-3d-gaussian
• convex-splatting-accepted-at-cvpr-cvpr2025 — convex-splatting-accepted-at-cvpr-cvpr2025
• dense-semantic-slam-based-on-2d-vision-language-foundation-models — dense-semantic-slam-based-on-2d-vision-language-foundation-models
• details-of-the-cvpr2025-award-candidate-papers-are-out-14 — details-of-the-cvpr2025-award-candidate-papers-are-out-14
• do-world-models-need-explicit-3d-our-answer-if-youre-using — do-world-models-need-explicit-3d-our-answer-if-youre-using
• do-you-want-to-create-gaussian-splatting-scenes-with-a-lot — do-you-want-to-create-gaussian-splatting-scenes-with-a-lot
• egg-fusion-efficient-3d-reconstruction-with-geometry-aware-g — egg-fusion-efficient-3d-reconstruction-with-geometry-aware-g
• egolifter-is-open-sourced-it-includes-full-recipe-for-project-aria-datasets-from — egolifter-is-open-sourced-it-includes-full-recipe-for-project-aria-datasets-from

Crypto (2)

• 1961042051669143681 — 전략형 오픈소스 트레이딩 봇 OctoBot
• 53m-total-gain-in-just-3-months-forget-gabagool22-this-is — 폴리마켓 봇 3개월 530만달러 수익

Dev-Tools (30)

• 1959480011691589708 — DeepCode 에이전트 코딩 프레임워크 급상승
• 1965646613747302663 — 도커 입문 서버 배포 가이드
• 30daymapchallenge-day-12-map-from-2125-step-into-2125-where — 30daymapchallenge-day-12-map-from-2125-step-into-2125-where
• 6-software-architectural-patterns---event-driven---layered---monolithic---micros — 소프트웨어 아키텍처 패턴 6선
• 673-released-extension-alert-viewcube-now-focuses-only-userl — Triplex 67.3·R3F VSCode 확장 업데이트
• a-curated-knowledge-bank-every-developer-wishes-they-had-sooner — 개발자 지식 뱅크 추천
• a-curated-list-of-awesome-cursorrules-files — 커서룰 추천 모음
• absolutely-fantastic-exposition-by-gordic_aleksa-a-great — GPU 행렬 곱셈 최적화 해설
• accidentally-landed-on-the-gpu-puzzles-book-from-modular-lik — Modular GPU Puzzles 학습서
• add-this-paragraph-to-the-file-to-turn-claude-code-into — Claude Teacher 문서화 프롬프트
• add-this-paragraph-to-the-httpstcocgo09sihd0-file-to-turn-claude-code-into-claud — Claude Teacher 프로젝트 문서화 루틴
• advanced-c-programming-perfect-for-anyone-who-wants-to-maste — 고급 C 학습 플레이리스트
• agentic-rag-for-stock-trading-if-you-want-a-fun-weekend-project-around-agentic-r — 주식 트레이딩 Agentic RAG 튜토리얼
• ai-engineering-관련-자료들-모음-chip-huyen이-ai-engineering-책을-쓰면서-참고했던-1200개-이상의-링크와-10 — AI Engineering 핵심 자료 큐레이션
• algorithms-for-modern-hardware — 현대 하드웨어 알고리즘 실전서
• also-check-the-introduction-to-parallel-computing-tutorial — 병렬 컴퓨팅 학습 자료
• article-memory-consumption-when-loading-uiimage-from-disk-by-antoine-a-uiimageco — UIImage 디스크 로드 메모리 최적화
• article-memory-consumption-when-loading-uiimage-from-disk-by-antoine-a — UIImage 디스크 로딩 메모리 전략
• attention-apple-developers-for-those-interested-in-meeting-the-apple-team-and-vi — 애플 쿠퍼티노 개발자 행사 모음
• attention-apple-developers-for-those-interested-in-meeting-the-apple-team-and — 애플 개발자센터 10월 이벤트
• blender-ai-motion-capture-plugin-supporting-real-time-capture-and-video-upload — 블렌더 AI 실시간 모션캡처 플러그인
• blender-alert-few-years-ago-said-on-blender-today-live-that-blender-could-come — 블렌더 iPad 네이티브 임박
• claude-code로-좋은-결과-얻기 — claude-code로-좋은-결과-얻기
• claude-cowork-활용-방식이야-개인마다-입맛에-맞게-anthropic에서-제안한-usecases — claude-cowork-활용-방식이야-개인마다-입맛에-맞게-anthropic에서-제안한-usecases
• collect-event-data-from-websites-and-apps-send-to-your-data — collect-event-data-from-websites-and-apps-send-to-your-data
• concurrency-in-webassembly-experiments-in-the-web-and-beyond-the-latest-acm — concurrency-in-webassembly-experiments-in-the-web-and-beyond-the-latest-acm
• crystal-upscaler-is-now-up-to-10-faster-upscaling-to-1k — crystal-upscaler-is-now-up-to-10-faster-upscaling-to-1k
• curated-list-of-tools-to-self-host-your-own-cloud-services — curated-list-of-tools-to-self-host-your-own-cloud-services
• deepwiki-어떤-코드베이스든-이해하기 — deepwiki-어떤-코드베이스든-이해하기
• each-week-gets-better-and-better-thanks-to-the-batteries-included-framework-for — each-week-gets-better-and-better-thanks-to-the-batteries-included-framework-for

Finance (3)

• 8개월간-5개의-llm에-14억원씩-맡겨-주식-거래를-시뮬레이션한-실험---gpt-5-claude-gemin — LLM 주식투자 모의실험
• ai-powered-hedge-fund-using-multiple-agents-for-trading-decisions — 멀티 에이전트 AI 헤지펀드 PoC
• algorithmic-trading-cheat-sheet-for-download — 알고리즘 트레이딩 공식 치트시트

GenAI (16)

• 1963580239541363098 — 뮌헨·런던 3D 파운데이션 모델 채용
• 1965495039217504464 — AI 자동 PBR 메쉬 텍스처링
• 2026-기술-트렌드-보고서-ai-에이전트부터-소버린-ai-피지컬-ai까지 — CB인사이트 2026 AI·테크 트렌드 14선
• 3dtopia-xl-for-pbr-asset-generation-is-accepted-to-cvpr2025 — 3DTopia-XL: Scaling High-quality 3D Asset Generation via Primitive Diffusion
• 3dtrajmaster-mastering-3d-trajectory-for-multi-entity-motion-in-video-generation — 3DTrajMaster: Mastering 3D Trajectory for Multi-Entity Motion in Video Generation
• 4dnex-feed-forward-4d-generative-modeling-made-easy — 4DNeX: Feed-Forward 4D Generative Modeling Made Easy
• a-deep-mathematical-exploration-of-diffusion-models-for-modern-ai-image-generati — Tutorial on Diffusion Models for Imaging and Vision
• added-confidence-aware-parallel-decoding-to-my-tiny-text-dif — 텍스트 Diffusion 확신 기반 병렬 디코딩
• after-reconx-further-explore-the-potential-of-video-diffusion-and-propose-dimens — DimensionX
• ai-just-made-try-before-you-buy-virtual — AI 가상 의류 시착
• alibaba-just-announced-uni3c-on-hugging-face-unifying — Uni3C 영상 생성 제어
• alibaba-presents-vace — 알리바바 VACE 영상 생성·편집
• animate3d-animating-any-3d-model-with-multi-view-video-diffusion-recent — Animate3D: Animating Any 3D Model with Multi-view Video Diffusion
• blenderfusion-3d-grounded-visual-editing-and-generative-compositing — BlenderFusion: 3D-Grounded Visual Editing and Generative Compositing
• code2video-a-code-centric-paradigm-for-educational-video — code2video-a-code-centric-paradigm-for-educational-video
• 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

LLM (13)

• 1961069820020961358 — LLM 앱 개발 학습 리소스
• 300명-사용자를-위한-자체-호스팅-llm-서버-구축이-가능할까요---reddit-rollama-서브레딧에-올라온-ᄌ — 300명-사용자를-위한-자체-호스팅-llm-서버-구축이-가능할까요---reddit-rollama-서브레딧에-올라온-ᄌ
• 5백만-건-이상의-문서를-처리하며-얻은-production-rag-경험---8개월간-rag검색-증강-생성 — 프로덕션 RAG 실전 교훈
• a-beautiful-work-dropped-on-arxiv — Foundations of Large Language Models
• a-new-framework-runs-mixtral-8x7b-at-11-tokenss-on-a-mobile — 모바일 Mixtral 8x7B 추론 가속
• a-very-hot-paper-just-dropped — LoRA vs Full Fine-tuning: An Illusion of Equivalence
• agentic-rag-for-stock-trading-if-you-want-a-fun-weekend-project-around-agentic — 주식 거래용 Agentic RAG 튜토리얼
• ai-engineer-interview-question — RAG 환각 개선 면접 질문
• ai가-사람을-인터뷰한다고요-내-취향-anthropic-interviewer-anthropic이-정말-재미있 — Anthropic Interviewer로 본 AI 업무 현실
• and-they-said-the-math-is-the-hardest-part-about-ml-httpstcog1fa80cokc — ML의 진짜 난관은 시스템
• anthropic-says-claude-doesnt-make-you-more-efficient-but-it- — How AI Impacts Skill Formation
• anthropic-says-claude-doesnt-make-you-more-efficient-but-it-does-make-you-stupid — How AI Impacts Skill Formation
• claude-code-성능-저하-추적용-일일-벤치마크 — Claude Code 일일 성능 벤치마크

Productivity (6)

• 9-years-to-become-a-doctor-but-6594-a-month-is-possible-with-chatgpt-in-19-minut — ChatGPT 월수입 만들기
• a-minimal-study-plan-for-machine-learning-python-maths-for-m — 머신러닝 학습 로드맵
• a-self-hostable-bookmark-everything-app-packed-with-ai-capabilities-ai-based-aut — AI 북마크 자동정리 앱 Karakeep
• ai-ml-roadmap-from-scratch — AI·ML 입문 로드맵
• ai-시대의-비판적-사고-critical-thinking---ai가-코드와-설계-답변을-만들어주는-환경에서도 — AI 시대 비판적 사고
• ask-hn-메모나-링크를-쌓아두기만-하고-실제-행동으로-옮기지-못하나요 — 메모 수집과 실행 병목

Rendering (45)

• 1960955497550635096 — Seam360GS: Seamless 360° Gaussian Splatting from Real-World Omnidirectional Images
• 1963887799100432386 — KIRI 4.0 스캔 PBR 자동화
• 1ngaussians-gt-differentiable-function-gt-mesh — 1ngaussians-gt-differentiable-function-gt-mesh
• 2次車に着手-1 — 2次車に着手-1
• 2次車に着手 — 2次車に着手
• 30fps-i-think-we-got-just-the-right-paper-for-you-give-us-a-week-for-arxiv-i-thi — 30fps-i-think-we-got-just-the-right-paper-for-you-give-us-a-week-for-arxiv-i-thi
• 3d-scene-camera-representation-with-joint-camera-photometric-optimization — 3D Scene-Camera Representation with Joint Camera Photometric Optimization
• 3dgeer-exact-and-efficient-volumetric-rendering-with-3d-gaussians-contributions — 3DGEER: Exact and Efficient Volumetric Rendering with 3D Gaussians
• 3dgeer-exact-and-efficient-volumetric-rendering-with-3d-gaussians-zixun-huang-ch — 3DGEER: Exact and Efficient Volumetric Rendering with 3D Gaussians
• 3dgrt-is-now-available-also-with-vulkan-api-as-part-of-the-vulkan-gaussian-splat — 3DGRT 벌칸 지원 추가
• 3dgsでリラリトできるのかすごいな — 언리얼 3DGS 리라이팅 플러그인
• 75k-words-releasing-in-1-3-days-enough-content-to-write-a-sm — 벌칸 엔진 개발기 공개 예고
• a-friend-with-a-goated-team-at-is-looking-for-a-cracked-frontend-grap — 4D 가우시안 스플래팅 프론트엔드 채용
• a-nice-course-on-inverse-graphics-from-light-transport-basic — 역그래픽스 강좌 추천
• a-nice-tutorial-of-gpu-raytracing-using-hip-through-orochi-from-a-simple — HIP/Orochi GPU 레이트레이싱 튜토리얼
• a-setup-to-add-points-to-a-curve-but-only-in-a-selected-area — 지오메트리 노드 곡선 선택 구간 포인트 추가
• a-whole-town-100m-gaussian-splats-rendered-in-real-time-with — 1억 가우시안 스플랫 도시 실시간 데모
• aaa-gaussians-anti-aliased-and-artifact-free-3d-gaussian — AAA-Gaussians: Anti-Aliased and Artifact-Free 3D Gaussian Rendering
• added-a-simple-benchmark-to-the-unity-webgpu-repository-1 — 유니티 WebGPU 벤치마크 추가
• added-a-simple-benchmark-to-the-unity-webgpu-repository — Unity WebGPU NoiseBall6 벤치마크
• added-the-ability-to-watch-the-rays-in-my-path-tracer-bounce-around-as-an-animat — 패스 트레이싱 레이 바운스 시각화
• adding-a-new-chapter-to-blender-geometry-nodes-series-on-youtube — 블렌더 지오메트리노드 책 더미 제작 튜토리얼
• adobe-research-introduces-v-rgbx-for-advanced-video-editing — V-RGBX: Video Editing with Accurate Controls over Intrinsic Properties
• all-thats-missing-now-is-some-ai-rendering-that-turns-the-retro-blocky-style — 블록풍→하이퍼리얼 AI 렌더링
• almost-5-hours-of-new-content-in-the-shader-course-just-covering-basic-particle — 셰이더 강좌 파티클 시스템 업데이트
• alpha-testing-the-new-3d-gaussian-splatting-paper-code — 와이드스크린 3DGS 렌더링 개선
• alright-my-webgpu-raytracer-is-completed-here-are-some-more-screenshots-live-lin — WebGPU 레이트레이서 완성 공개
• and-here-are-the-slides-from-my-game-engine-architecture-talk-at-virginia-tech-l — 게임 엔진 아키텍처 개요
• at-the-legendary-siggraph-advances-in-real-time-rendering-course-today-activison — 워존 칼데라 USD 데이터셋 공개
• at-the-legendary-siggraph-advances-in-real-time-rendering-course-today — 워존 칼데라 USD 데이터셋 공개
• awesome-example-of-relighting-3dgs-scans-in-blender-using-the-3dgs-render-addon — 블렌더 3DGS 리라이팅 워크플로
• beeble-just-rolled-out-switchlight-20-its-new-video-to-pbr-ai-model-the-tech — SwitchLight 2.0 영상 재조명
• blender-50-cycles-now-supports-thin-film-iridescence-for — Blender 5.0 Cycles 박막 무지개빛
• blog-post-cheap-ray-tracing-using-only-acceleration-structur — 가속 구조 기반 미니멀 레이 트레이싱
• btw-the-project-is-completely-open-source-and-fairly-readable — WebGPU 오픈소스 레이트레이서
• built-this-installation-for-jr-with-the-webgpurender-of-thre — built-this-installation-for-jr-with-the-webgpurender-of-thre
• check-out-s-paper-github-and-tools-paper-code-mit-license-de — check-out-s-paper-github-and-tools-paper-code-mit-license-de
• check-out-the-latest-radiance-field-representation — check-out-the-latest-radiance-field-representation
• clift-compressive-light-field-tokens-for-compute-efficient-and-adaptive-neural — clift-compressive-light-field-tokens-for-compute-efficient-and-adaptive-neural
• code-dropped — code-dropped
• cool-work — cool-work
• cuda-ray-tracing-36x-faster-than-rtx-my-cuda-ray-tracing-journey-introduction — cuda-ray-tracing-36x-faster-than-rtx-my-cuda-ray-tracing-journey-introduction
• did-a-4k-texture-upload-test-using-amp-on-apple-silicon-m4-max-safari-is-3070 — did-a-4k-texture-upload-test-using-amp-on-apple-silicon-m4-max-safari-is-3070
• did-a-small-comparison-of-babylonjs-playcanvas-and-threejs — did-a-small-comparison-of-babylonjs-playcanvas-and-threejs
• diffuman4d-4d-consistent-human-view-synthesis-from-sparse-view-videos-with — diffuman4d-4d-consistent-human-view-synthesis-from-sparse-view-videos-with

Robotics (20)

• 1965106627511546357 — Pre-training Auto-regressive Robotic Models with 4D Representations
• 1x-world-model-from-video-to-action-a-new-way-robots-learn — 1x-world-model-from-video-to-action-a-new-way-robots-learn
• 3d-gaussian-splatting-in-robotics-a-survey-paper — 3d-gaussian-splatting-in-robotics-a-survey-paper
• 4dgs-slam — 4D Gaussian Splatting SLAM
• a-compact-04b-vision-language-action-model-that-finally-lets-robots-manipulate-m — DynamicVLA
• a-must-read-survey-efficient-vision-language-action-models-i — A Survey on Efficient Vision-Language-Action Models
• a-new-30-minute-presentation-from-aelluswamy-teslas-vp-of — 테슬라 FSD·AI 진행 발표
• a-taste-of-teleoperation-nothing-beats-getting-hands-on-with — 로봇 텔레오퍼레이션 실습
• accepted-theia-distilling-diverse-vision-foundation-models-for-robot-learning-th — Theia: Distilling Diverse Vision Foundation Models for Robot Learning
• added-more-arms-the-entire-setup-is-now-about-600 — 저비용 다중 로봇팔 세팅
• after-years-of-research-there-is-finally-a-solution-to-long — TTT-E2E 로봇 장문맥 메모리
• awesome-robotics-3d-a-curative-list-of-3d-vision-papers-relating-to-robotics-dom — Awesome-Robotics-3D 논문 큐레이션
• awesome-robotics-3d-a-curative-list-of-3d-vision-papers-relating-to-robotics — 로보틱스 3D 비전 Awesome 리스트
• can-a-robot-inspect-all-views-of-an-object-today-iros-we — can-a-robot-inspect-all-views-of-an-object-today-iros-we
• can-robots-search-for-objects-like-humans-humans-explore-unseen-environments — can-robots-search-for-objects-like-humans-humans-explore-unseen-environments
• can-we-bridge-the-sim-to-real-gap-in-complex-manipulation-wi-683188 — can-we-bridge-the-sim-to-real-gap-in-complex-manipulation-wi-683188
• check-out-our-low-cost-3d-printable-exoskeleton-system-that-can-teleoperate-many — check-out-our-low-cost-3d-printable-exoskeleton-system-that-can-teleoperate-many
• check-out-our-low-cost-3d-printable-exoskeleton-system-that-can-teleoperate — check-out-our-low-cost-3d-printable-exoskeleton-system-that-can-teleoperate
• cyberdemo-is-accepted-at-check-it-out — cyberdemo-is-accepted-at-check-it-out
• dynamicvla — dynamicvla

Simulation (10)

• 1961158594495484037 — Omniverse NuRec AV 재구성
• 1965611185971015864 — Vid2Sim: Realistic and Interactive Simulation from Video for Urban Navigation
• 3d-digital-twin-of-a-new-light-rail-project-in-australia — 3d-digital-twin-of-a-new-light-rail-project-in-australia
• 3d-gaussian-splatting-driving-simulator-you-know-those — 3d-gaussian-splatting-driving-simulator-you-know-those
• aircraft-design-with-physics-ai-rune-aero-is-transforming — AI 가상 풍동 항공기 설계
• always-wanted-to-play-with-mlx-so-heres-my-first-attempt-potential-energy-minimi — Swift MLX 스프링 에너지 최소화
• can-we-learn-a-3d-world-model-that-predicts-object-dynamics-directly-from — can-we-learn-a-3d-world-model-that-predicts-object-dynamics-directly-from
• check-out-vid2sim-generalizable-video-based-reconstruction-of-appearance — check-out-vid2sim-generalizable-video-based-reconstruction-of-appearance
• checkout-our-latest-work-on-gaussian-splatting-for-lidar — checkout-our-latest-work-on-gaussian-splatting-for-lidar
• 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

VLM (7)

• a-new-highly-accurate-ocr-was-just-released-and-its-open-source-surya-is-accurat — 오픈소스 고정밀 OCR Surya 공개
• advice-for-ai-engineers-a-small-visual-language-model-fine — 소형 VLM 파인튜닝 비용 효율
• alpha-clip-a-clip-model-focusing-on-wherever-you-want-paper-page-contrastive-lan — Alpha-CLIP: A CLIP Model Focusing on Wherever You Want
• an-ocr-web-app-to-extract-find-and-describe-text — OCR 웹앱 텍스트 추출
• booom-learn-vlms-from-inside-out-in-lt-1000-lines-of-pure — 1000줄 파이토치 nanoVLM 학습
• can-llm-understand-and-reason-explicit-object-coordinates-introducing-pixelllm-w — can-llm-understand-and-reason-explicit-object-coordinates-introducing-pixelllm-w
• can-vlms-really-think-in-4d-3d-space-time-when-a-model-cant — can-vlms-really-think-in-4d-3d-space-time-when-a-model-cant

XR (16)

• 2100-what-if-your-hands-became-the-controllers — 2100-what-if-your-hands-became-the-controllers
• 3d-capture-is-dope-for-media-entertainment-but-the-utilitari — 3d-capture-is-dope-for-media-entertainment-but-the-utilitari
• 3d-hand-tracking-on-smart-glasses-not-only-can-we-stream — 3d-hand-tracking-on-smart-glasses-not-only-can-we-stream
• 4100-one-thing-i-realized-while-working-in-mixed-reality-pro — 퀘스트 3 MR 적응형 조명 도구
• a-great-visual-positioning-system-makes-augmented-reality-fe — 야간 온디바이스 시각 위치추정
• add-sphere-apply-color-yellow-clone-bit-boring-to-create-a-w — Vision Pro WebXR 음성 객체 생성
• after-i-shared-my-prototype-with-an-ar-camera-portal-a-few — 물리 연동 AR 포털 프로토타입
• ai-ar-is-going-to-change-everything — Vision Pro 3D 보블헤드 제작
• all-code-data-and-models-are-now-released-for-simxr-1 — SimXR 코드·데이터·모델 공개
• all-code-data-and-models-are-now-released-for-simxr — SimXR: Real-Time Simulated Avatar from Head-Mounted Sensors
• android-xr-quick-facts-without-visionoss-unified-rendering-which-lets-multitaski — Android XR Home/Full Space 차이
• apple-seems-to-have-some-pretty-good-visual-odometry-under-t — 비전 프로 시각 오도메트리 성능
• as-an-alternative-to-reality-converter-beta-i-use-two-conversion-flows-with-blen — Blender USDZ 변환 워크플로
• as-an-alternative-to-reality-converter-beta-i-use-two-conversion-flows-with — Blender 기반 USDZ 변환 플로우
• career-update-happy-to-share-that-ive-joined-meta-reality-labs-as-a-research — career-update-happy-to-share-that-ive-joined-meta-reality-labs-as-a-research
• each-week-3dstreetapp-gets-better-and-better-thanks-to-the-batteries-included — each-week-3dstreetapp-gets-better-and-better-thanks-to-the-batteries-included

기타 (6)

• 2026 — 2026
• a-2196-page-free-book-released-in-feb-2024-algebra-topology — 무료 2196쪽 ML 수학 교재
• a-680-page-interactive-book-on-computer-science-algorithms-this-might-be-the — 인터랙티브 CS 알고리즘 책
• a-really-simple-mathematical-blog-explaining — 시간 역전파(BPTT) 기울기 계산
• a-simple-beautiful-blog-on-understanding-eigenvectors-and — 고유벡터·고유값 시각적 이해
• corvids-are-known-to-be-smart-problem-solvers-this-one-for-e — corvids-are-known-to-be-smart-problem-solvers-this-one-for-e