Tootfinder

Opt-in global Mastodon full text search. Join the index!

No exact results. Similar results found.
@leftsidestory@mstdn.social
2025-08-13 05:44:27

City Features 🌆
城市特征 🌆
📷 Pentax MX
🎞️Ilford FP4 Plus, expired 1994
buy me ☕️ ?/请我喝杯☕️？
#filmphotography

Ilford FP4 Plus 125 (FF) • English Alt Text: The image shows the exterior of a building with a row of stools lined up outside. The building has several arched windows with awnings, and there is a glass door on the right side. There are signs with text in Chinese on the wall next to the door. The scene appears to be on a quiet street with other buildings visible in the background. • Chinese Alt Text: 这张图片展示了一栋建筑的外部，外面排着一排凳子。建筑有几个带遮阳篷的拱形窗户，右侧有一扇玻璃门。门旁的墙上有一些中文标志。场景似乎在一条安静的街道上，背景中可以看到其他建筑。
Ilford FP4 Plus 125 (FF) • English Alt Text: A nighttime cityscape featuring several tall buildings with illuminated windows. Prominent advertisements are visible on the buildings, including “58同城,” “58集团,” and “pwc 普华永道.” The architecture creates a striking urban composition. • Chinese Alt Text: 夜间城市景观，画面中有几栋高楼，窗户灯光明亮。建筑上可见显眼的广告标语，包括“58同城”、“58集团”和“pwc 普华永道”。整体构图展现了都市建筑的独特美感。
Ilford FP4 Plus 125 (FF) • English Alt Text: A black-and-white photo of an outdoor urban space, framed by a dark structure in the foreground. The scene includes a large planter with tall grass, a checkered pavement, trees, pedestrians, a car, and dome-shaped architectural structures in the background. • Chinese Alt Text: 一张黑白照片，展示了一个户外城市空间，前景由一个深色结构框住。画面中有一个种满高草的大型花盆、棋盘格铺地、树木、行人、一辆汽车，以及背景中的圆顶建筑和独特设计的楼房。
Ilford FP4 Plus 125 (FF) • English Alt Text: A plastic bottle sits atop a structure covered in small, square mosaic tiles. The bottle contains small objects, possibly beads or pellets. The background is dark and indistinct, suggesting a nighttime or shaded setting. • Chinese Alt Text: 一个塑料瓶放在一个覆盖着小方形马赛克瓷砖的结构上。瓶子里装有一些小物体，可能是珠子或颗粒。背景昏暗模糊，似乎是在夜间或阴影处拍摄。
@tiotasram@kolektiva.social
2025-08-15 03:27:16

US ethnic cleansing and what to do about it
Reposting link to source article instead of screenshot of tweet that had no alt text:
Data on arrests shows that ICE was heavily engaged in racial profiling in LA, because their arrest numbers fell by ~66% after that were ordered to stop making arrests based just in factors like skin color, with place, or language spoken.
#ICE #USPol

@memeorandum@universeodon.com
2025-09-13 18:41:15

Tyler Robinson's roommate who tipped off FBI is identified as family refuse to deny transgender motive for Charlie Kirk assassination (Josh Boswell/Daily Mail)
dailymail.co.uk/news/article-1
memeorandum.com/250913/p39#a25

@arXiv_csCL_bot@mastoxiv.page
2025-08-14 09:47:12

UtterTune: LoRA-Based Target-Language Pronunciation Edit and Control in Multilingual Text-to-Speech
Shuhei Kato
arxiv.org/abs/2508.09767 ar…

UtterTune: LoRA-Based Target-Language Pronunciation Edit and Control in Multilingual Text-to-Speech
We propose UtterTune, a lightweight adaptation method that fine-tunes a multilingual text-to-speech (TTS) system based on a large language model (LLM) architecture, designed to enhance the controllability of pronunciation in a target language while preserving performance in others. While LLM architectures have enabled TTS models to achieve remarkable naturalness, accurately modeling grapheme-to-phoneme (G2P) mapping and prosody remains challenging, especially when the model omits an explicit G2…

@arXiv_csCV_bot@mastoxiv.page
2025-07-14 10:08:32

Lumos-1: On Autoregressive Video Generation from a Unified Model Perspective
Hangjie Yuan, Weihua Chen, Jun Cen, Hu Yu, Jingyun Liang, Shuning Chang, Zhihui Lin, Tao Feng, Pengwei Liu, Jiazheng Xing, Hao Luo, Jiasheng Tang, Fan Wang, Yi Yang
arxiv.org/abs/2507.08801

Lumos-1: On Autoregressive Video Generation from a Unified Model Perspective
Autoregressive large language models (LLMs) have unified a vast range of language tasks, inspiring preliminary efforts in autoregressive video generation. Existing autoregressive video generators either diverge from standard LLM architectures, depend on bulky external text encoders, or incur prohibitive latency due to next-token decoding. In this paper, we introduce Lumos-1, an autoregressive video generator that retains the LLM architecture with minimal architectural modifications. To inject s…

@arXiv_csSD_bot@mastoxiv.page
2025-08-14 09:16:52

Analysis of Domain Shift across ASR Architectures via TTS-Enabled Separation of Target Domain and Acoustic Conditions
Tina Raissi, Nick Rossenbach, Ralf Schl\"uter
arxiv.org/abs/2508.09868

Analysis of Domain Shift across ASR Architectures via TTS-Enabled Separation of Target Domain and Acoustic Conditions
We analyze automatic speech recognition (ASR) modeling choices under domain mismatch, comparing classic modular and novel sequence-to-sequence (seq2seq) architectures. Across the different ASR architectures, we examine a spectrum of modeling choices, including label units, context length, and topology. To isolate language domain effects from acoustic variation, we synthesize target domain audio using a text-to-speech system trained on LibriSpeech. We incorporate target domain n-gram and neural …

@leftsidestory@mstdn.social
2025-07-14 00:30:06

Weird Greco 🇬🇷
怪异希腊 🇬🇷
📷 Nikon FE
🎞️FOMAPAN Action 400
buy me ☕️ ?/请我喝杯☕️？
#filmphotography

Foma FOMAPAN 400 Action (FF) 🌿 English alt text: A black and white photograph of a garden path lined with lush, overgrown foliage. In the distance, a white column supports a small structure, possibly a gazebo. Two people stand nearby, partially hidden by the plants, evoking a sense of quiet retreat. 🌿 中文替代文字： 一张黑白照片，展示了一条被茂密植物包围的花园小径。远处有一根白色柱子支撑着一个可能是凉亭的结构，两个人站在旁边，被植物部分遮挡，营造出宁静隐逸的氛围。
Foma FOMAPAN 400 Action (FF) 🏛️ English alt text: A black and white photograph of a statue featuring two classical figures. One figure stands with an arm raised, holding a torch-like object, while the other mirrors the pose with theirs lowered. The statue is crafted from stone or marble and is framed by palm trees and a tall, modern building with grid-like windows. The contrasting backdrop blends timeless artistry with urban architecture. 🏛️ 中文替代文字： 一张黑白照片，展示了一个由石材或大理石雕刻而成的雕像，包含两位古典人物。其中一人高举手…
Foma FOMAPAN 400 Action (FF) 🕊️ English alt text: A detailed fountain featuring two winged angel-like figures sitting back-to-back on a sculptural base. Their wings are gracefully extended, and their robes flow naturally over the contours of the fountain. Surrounded by neatly trimmed hedges and a tiled water basin, the scene exudes calm elegance in an outdoor garden setting. 🕊️ 中文替代文字： 一个精致的喷泉，底座上背靠背坐着两位带翅膀的天使般雕像，羽翼优雅展开，袍服自然垂落，雕刻精美。喷泉周围环绕着修剪整齐的绿篱和瓷砖水池，整体环境宁静而高雅，仿佛置身于一处花园中的艺术角落。
Foma FOMAPAN 400 Action (FF) 🌴 English alt text: A black and white photograph of a multi-story Mediterranean-style residential building with tiled rooftops and arched windows. In the foreground, several tall palm trees dominate the scene, their fronds reaching upward and adding contrast to the architectural geometry. The image evokes a quiet coastal or resort-town atmosphere, where nature and human design coexist peacefully. 🌴 中文替代文字： 一张黑白照片，画面中心是一栋多层的地中海风格住宅建筑，屋顶铺设瓷砖，窗户呈弧形。前景中有几棵高大的棕榈树，树叶向上伸…
@fgraver@hcommons.social
2025-09-10 19:22:24

Pay-per-output? AI firms blindsided by beefed up robots.txt instructions. arstechnica.com/tech-policy/20

Pay-per-output? AI firms blindsided by beefed up robots.txt instructions.
“Really Simple Licensing” makes it easier for creators to get paid for AI scraping.

@arXiv_csCL_bot@mastoxiv.page
2025-08-13 10:17:12

Utilizing Multilingual Encoders to Improve Large Language Models for Low-Resource Languages
Imalsha Puranegedara, Themira Chathumina, Nisal Ranathunga, Nisansa de Silva, Surangika Ranathunga, Mokanarangan Thayaparan
arxiv.org/abs/2508.09091

Utilizing Multilingual Encoders to Improve Large Language Models for Low-Resource Languages
Large Language Models (LLMs) excel in English, but their performance degrades significantly on low-resource languages (LRLs) due to English-centric training. While methods like LangBridge align LLMs with multilingual encoders such as the Massively Multilingual Text-to-Text Transfer Transformer (mT5), they typically use only the final encoder layer. We propose a novel architecture that fuses all intermediate layers, enriching the linguistic information passed to the LLM. Our approach features tw…

@leftsidestory@mstdn.social
2025-09-14 00:30:01

On The Road - To Xi’An/Before & After 🌬️
在路上 - 去西安/之前之后🌬️
📷 Pentax MX
🎞️Lucky SHD 400
#filmphotography #Photography #blackandwhite

Lucky Lucky SHD 400 (FF) English Alt Text: A black-and-white urban photo captures a modern high-rise apartment building with balconies and clean lines. Behind it stands a more industrial-looking structure with a grid-like facade and exposed upper section. In the foreground, a pedestrian bridge or overpass with metal railings crosses the scene. The architectural contrast and geometric shapes give the image a stark, modern aesthetic. 中文替代文本： 这是一张黑白城市照片，画面中是一栋现代高层公寓楼，带有阳台和简洁的线条。其后是一座工业风格的建筑，外墙呈网…
Lucky Lucky SHD 400 (FF) English Alt Text: A nighttime cityscape in black and white features several buildings. On the left, tall residential or office towers with lit windows rise into the dark sky. In the foreground, a low building with utility lines adds depth. On the right, a darker structure displays two glowing circular clock faces, suggesting a clock tower. The contrast between light and shadow creates a quiet, contemplative mood. 中文替代文本： 这是一张黑白城市夜景照片。左侧是几栋高层住宅或办公楼，窗户中透出灯光，直指夜空。前景中有一座低…
Lucky Lucky SHD 400 (FF) English Alt Text: A black-and-white photograph shows a wire basket filled with handheld fans. The fans are arranged with their handles pointing downward and their circular heads fanning outward like petals. Each fan has a mesh or slatted surface, and they vary slightly in shape and size. The basket rests on a smooth indoor floor, possibly in a communal or recreational space. The background is softly blurred, with indistinct furniture and objects. The monochrome tones a…