So like if you provide guidelines, instructions, process on how you write custom parsing, etc that would be really helpful
Ujjwal Tyagi
Ujjwal-Tyagi
AI & ML interests
Chief Scientist at Shirova AI, Senior ML Engineer and Computer Scientist. Passionate about open-source AI, with AI ethics and safety at the heart of everything I build.
Recent Activity
published
a model about 2 hours ago
Ujjwal-Tyagi/Empty-500B-Base-MoE liked
a model about 3 hours ago
Qwen/Qwen3.5-35B-A3B-Base liked
a dataset about 4 hours ago
agents-course/course-certificates-of-excellence Organizations
We can't trust Anthropic, even they have stolen pirated data from Lib Genesis and thousands of copyrighted songs, our data is not safe
replied to AbstractPhil's post about 24 hours ago
Oh wow, looks interesting
So like, It can be good for training training vision language models, also helpful for robotics...too and also good for training text to image models, nice work! I wanted to know that how you collect these kind of amazing datasets?
reacted to nyuuzyou's post with 👍🔥 1 day ago
Post
1751
🌍 Street-Level Imagery Dataset nyuuzyou/streetview
934,191 image records index Eastern Europe and Northern Asia. Temporal links map historical views at identical coordinates across nine years.
Key Stats:
- 905,940 unique images
- Coverage spanning 2016 to 2025
- Average 14.3 historical links per location
Geographic bounds span 20.49° E to 152.32° E. Urban centers show higher data density.
934,191 image records index Eastern Europe and Northern Asia. Temporal links map historical views at identical coordinates across nine years.
Key Stats:
- 905,940 unique images
- Coverage spanning 2016 to 2025
- Average 14.3 historical links per location
Geographic bounds span 20.49° E to 152.32° E. Urban centers show higher data density.
-
- 3 replies
replied to SeaWolf-AI's post 1 day ago
Glad to see this amazing work
reacted to their post with 😎 4 days ago
Post
2770
Public reports allege that Anthropic gobbled up trillions of tokens of copyrighted material and public data to build their castle. 🏰📄 Now that they're sitting on top, they're begging for special laws to protect their profits while pulling the ladder up behind them. 🪜🚫
But the hypocrisy meter just broke! 📉 They are accusing Chinese labs like DeepSeek, Minimax, and Kimi of "huge distillation attacks. The Reality is that You can't just loot the entire internet's library, lock the door, and then sue everyone else for reading through the window. Stop trying to gatekeep the tech you didn't own in the first place. Read the complete article on it: https://huggingface.co/blog/Ujjwal-Tyagi/the-dark-underbelly-of-anthropic
But the hypocrisy meter just broke! 📉 They are accusing Chinese labs like DeepSeek, Minimax, and Kimi of "huge distillation attacks. The Reality is that You can't just loot the entire internet's library, lock the door, and then sue everyone else for reading through the window. Stop trying to gatekeep the tech you didn't own in the first place. Read the complete article on it: https://huggingface.co/blog/Ujjwal-Tyagi/the-dark-underbelly-of-anthropic
-
- 3 replies
replied to their post 5 days ago
I actually think about these memes 😂 🤣
replied to their post 5 days ago
But it's true that Moonshot AI did heavy distillation of Claude models to build Kimi K2.5, as if you ask Kimi-K2.5 that "who are you" it says "I am Claude built by Anthropic", Anthropic is trying to protect their profit but they are quite right that about the safety of the community because there is no trust for Chinese AI Companies
posted an
update 5 days ago
Post
2770
Public reports allege that Anthropic gobbled up trillions of tokens of copyrighted material and public data to build their castle. 🏰📄 Now that they're sitting on top, they're begging for special laws to protect their profits while pulling the ladder up behind them. 🪜🚫
But the hypocrisy meter just broke! 📉 They are accusing Chinese labs like DeepSeek, Minimax, and Kimi of "huge distillation attacks. The Reality is that You can't just loot the entire internet's library, lock the door, and then sue everyone else for reading through the window. Stop trying to gatekeep the tech you didn't own in the first place. Read the complete article on it: https://huggingface.co/blog/Ujjwal-Tyagi/the-dark-underbelly-of-anthropic
But the hypocrisy meter just broke! 📉 They are accusing Chinese labs like DeepSeek, Minimax, and Kimi of "huge distillation attacks. The Reality is that You can't just loot the entire internet's library, lock the door, and then sue everyone else for reading through the window. Stop trying to gatekeep the tech you didn't own in the first place. Read the complete article on it: https://huggingface.co/blog/Ujjwal-Tyagi/the-dark-underbelly-of-anthropic
-
-
- 3 replies
Cool
Glad to see, wow
replied to SeaWolf-AI's post 7 days ago
Glad to see this benchmark, good work
reacted to SeaWolf-AI's post with 👍😎❤️🤝🤗🔥 7 days ago
Post
4209
FINAL Bench Released: The Real Bottleneck to AGI Is Self-Correction
We release FINAL Bench, the first benchmark for measuring functional metacognition in LLMs — the ability to detect and correct one's own reasoning errors. Every existing benchmark measures final-answer accuracy. None measures whether AI knows it is wrong.
Dataset: [FINAL-Bench/Metacognitive]( FINAL-Bench/Metacognitive) | 100 Tasks | 15 Domains | 8 TICOS Types | Apache 2.0
Leaderboard: FINAL-Bench/Leaderboard
Article: https://huggingface.co/blog/FINAL-Bench/metacognitive
Core Innovation
Our 5-axis rubric separates what no prior benchmark could: MA (Metacognitive Accuracy) — the ability to say "I might be wrong", and ER (Error Recovery) — the ability to actually fix it. This maps directly to the monitoring-control model of Nelson & Narens (1990) in cognitive psychology.
Three Findings Across 9 SOTA Models
We evaluated GPT-5.2, Claude Opus 4.6, Gemini 3 Pro, DeepSeek-V3.2, Kimi K2.5, and others across 100 expert-level tasks:
1. ER Dominance. 94.8% of MetaCog gain comes from Error Recovery alone. The bottleneck to AGI is not knowledge or reasoning — it is self-correction.
2. Declarative-Procedural Gap. All 9 models can verbalize uncertainty (MA = 0.694) but cannot act on it (ER = 0.302). They sound humble but fail to self-correct — the most dangerous AI safety profile.
3. Difficulty Effect. Harder tasks benefit dramatically more from metacognition (Pearson r = -0.777, p < 0.001).
Paper: FINAL Bench: Measuring Functional Metacognitive Reasoning in LLMs
FINAL Bench is the first tool to tell apart what AI truly knows from what it merely pretends to know.
We release FINAL Bench, the first benchmark for measuring functional metacognition in LLMs — the ability to detect and correct one's own reasoning errors. Every existing benchmark measures final-answer accuracy. None measures whether AI knows it is wrong.
Dataset: [FINAL-Bench/Metacognitive]( FINAL-Bench/Metacognitive) | 100 Tasks | 15 Domains | 8 TICOS Types | Apache 2.0
Leaderboard: FINAL-Bench/Leaderboard
Article: https://huggingface.co/blog/FINAL-Bench/metacognitive
Core Innovation
Our 5-axis rubric separates what no prior benchmark could: MA (Metacognitive Accuracy) — the ability to say "I might be wrong", and ER (Error Recovery) — the ability to actually fix it. This maps directly to the monitoring-control model of Nelson & Narens (1990) in cognitive psychology.
Three Findings Across 9 SOTA Models
We evaluated GPT-5.2, Claude Opus 4.6, Gemini 3 Pro, DeepSeek-V3.2, Kimi K2.5, and others across 100 expert-level tasks:
1. ER Dominance. 94.8% of MetaCog gain comes from Error Recovery alone. The bottleneck to AGI is not knowledge or reasoning — it is self-correction.
2. Declarative-Procedural Gap. All 9 models can verbalize uncertainty (MA = 0.694) but cannot act on it (ER = 0.302). They sound humble but fail to self-correct — the most dangerous AI safety profile.
3. Difficulty Effect. Harder tasks benefit dramatically more from metacognition (Pearson r = -0.777, p < 0.001).
from datasets import load_dataset
dataset = load_dataset("FINAL-Bench/Metacognitive", split="train")Paper: FINAL Bench: Measuring Functional Metacognitive Reasoning in LLMs
FINAL Bench is the first tool to tell apart what AI truly knows from what it merely pretends to know.
-
- 6 replies