Jingkang Yang @NTU🇸🇬
@JingkangY
Followers
2K
Following
1K
Media
67
Statuses
351
NTU MMLab PhD Student - Reasoning in the Open World. ECCV’22 Best Backpack Award 🎒
Singapore
Joined March 2021
RT @cwolferesearch: The gpt-oss models from OpenAI are a synthesis of ideas from prior research. Here are 10 interesting papers that were d….
0
93
0
RT @AtsuMiyaiAM: #ACL2025 .I'll present our work, Unsolvable Problem Detection (, Today! .If you're interested, fee….
0
9
0
RT @BoLi68567011: SAE Made Easy. Sparse Autoencoders (SAE) have become a cornerstone in the field of explainable A….
0
6
0
RT @KaichenZhang358: Tired of hooking SAEs to different models?. Check out our new repo for plug-and-play SAE training—now as easy as other….
github.com
A framework that allows you to apply Sparse AutoEncoder on any models - EvolvingLMMs-Lab/sae
0
4
0
RT @AtsuMiyaiAM: 🎉Our survey on how OOD detection & related tasks have evolved in the VLM and Large VLM era is accepted to #TMLR!. The fiel….
0
24
0
RT @RuiqisNotes: 🎉 Excited to share our new work: 👓Ego-R1!. We cracked ultra-long egocentric video reasoning! 🤯 Think days/weeks of footage….
0
8
0
RT @richardzhangsfu: My first "Ego" involvement 🤣 Ego-vision, Ego3D/4D are at the heart of affordance discovery, interaction exploration, a….
0
5
0
RT @shulin_tian: 🎥 Video is already a tough modality for reasoning. Egocentric video? Even tougher! It is longer, messier, and harder. 💡 H….
0
9
0
RT @Ar_Douillard: Ego-R1: Chain-of-Tool-Thought for Ultra-Long Egocentric Video Reasoning. Reasoning over weeks-long video, with a mix of C….
0
9
0
RT @liuziwei7: 🔥Multi-turn Grounding-based Policy Optimization (MGPO)🔥. MGPO equips LMMs with o3-style interpretable, multi-turn visual gro….
0
37
0
RT @TheAITalksOrg: 𝕋𝕙𝕖 𝕌𝕡𝕔𝕠𝕞𝕚𝕟𝕘 𝔸𝕀 𝕋𝕒𝕝𝕜: .Towards Packing the Intelligence of Large Foundation Models on Low-Resource Devices .by Dr. Sourv….
0
2
0
RT @AtsuMiyaiAM: 🤔 Multiple-choice questions are a common benchmark format, but do LMMs really understand the answers?. 📣 In our #ACL2025 (….
0
28
0
RT @liuziwei7: Glad to see that Gemini 2.5 pro achieves impressive performance on our college-level video reasoning benchmark 📊Video-MMMU📊….
0
9
0
RT @liuziwei7: Congrats to Jingkang @JingkangY and Lingdong @ldkong1205 for being selected as #CVPR2025 Outstanding Reviewers!.
0
4
0
🎧 Introducing Aero-1-Audio. 📢📢📢 1.5B audio-language model from the Aero-1 series by @lmmslab!. Checkout our blog -
lmms-lab.com
Aero-1-Audio is a 1.5B compact audio model capable of handling a range of audio tasks, including speech recognition, audio understanding, and audio instructions following.
Aero-1-Audio is out on Hugging Face. Trained in <24h on just 16×H100.Handles 15+ min audio seamlessly .Outperforms bigger models like Whisper, Qwen-2-Audio & commercial services from ElevenLabs/Scribe
1
12
49
RT @_akhaliq: Aero-1-Audio is out on Hugging Face. Trained in <24h on just 16×H100.Handles 15+ min audio seamlessly .Outperforms bigger mod….
0
64
0
RT @BoLi68567011: 🚀 Introducing Aero-1-Audio — a compact yet mighty audio model. ⚡ Trained in <24h on just 16×H100.🎧 Handles 15+ min audio….
0
8
0
🎧 Introducing Aero-1-Audio.A 1.5B audio-language model from the Aero-1 series by @lmmslab!. The model is trained in 1 day on 50K hrs of curated data. It can handle 15 minutes of continuous audio without splitting. Blog: HF Demo:
huggingface.co
🚀 Introducing Aero-1-Audio — a compact yet mighty audio model. ⚡ Trained in <24h on just 16×H100.🎧 Handles 15+ min audio seamlessly.💡 Outperforms bigger models like Whisper, Qwen-2-Audio & commercial services from ElevenLabs/Scribe. Aero shows: smart data > massive scale.
0
1
9
RT @BoLi68567011: # LMMs-Eval v0.3.3 Updates. 🌟 We added `sglang_srt`, `vllm` and `openai_compatible` models to support wide range opensour….
0
9
0