LiquidAI/LFM2.5-2.6B
Text Generation ā¢ 3B ā¢ Updated ā¢ 93.7k ā¢ 570
Lian Junhong
THEATLAS
AI & ML interests
Natural language processing (NLP), personalized generation in large language models (LLMs), and applications of NLP in the financial domain.
Recent Activity
liked a model 2 days ago
LiquidAI/LFM2.5-2.6B posted an update 2 months ago
š¤ We train LLMs ā but how do LLMs reshape AI research itself?
Sharing our new survey, AI4AIR: A Comprehensive Survey on Large Language Models for AI Research š
Rather than treating LLMs as outside helpers for writing or literature review, we map how they enter the ML pipeline itself ā through a 2D taxonomy (research domain Ć pipeline stage) and 5 recurring roles:
š·ļø Annotator Ā· š§Ŗ Synthesizer Ā· āļø Optimizer Ā· š Evaluator Ā· šļø Orchestrator
Our key takeaway: how deeply an LLM can empower a stage is bounded by its validation cost ā which is exactly why data annotation & evaluation are already mature, while model design & full-loop orchestration remain hard.
š Paper (PDF): https://github.com/ICT-FinD-Lab/Awesome-LLMs-for-AI-Research/blob/master/PDF/AI4AIR_Survey_v260601.pdf
š Project: https://ict-find-lab.github.io/Awesome-LLMs-for-AI-Research
ā GitHub: https://github.com/ICT-FinD-Lab/Awesome-LLMs-for-AI-Research
Stars & feedback warmly welcome š upvoted a paper 7 months ago
GDCNet: Generative Discrepancy Comparison Network for Multimodal Sarcasm DetectionOrganizations
liked a
model 2 days ago
posted an update 2 months ago
Post
96
š¤ We train LLMs ā but how do LLMs reshape AI research itself?
Sharing our new survey, AI4AIR: A Comprehensive Survey on Large Language Models for AI Research š
Rather than treating LLMs as outside helpers for writing or literature review, we map how they enter the ML pipeline itself ā through a 2D taxonomy (research domain Ć pipeline stage) and 5 recurring roles:
š·ļø Annotator Ā· š§Ŗ Synthesizer Ā· āļø Optimizer Ā· š Evaluator Ā· šļø Orchestrator
Our key takeaway: how deeply an LLM can empower a stage is bounded by its validation cost ā which is exactly why data annotation & evaluation are already mature, while model design & full-loop orchestration remain hard.
š Paper (PDF): https://github.com/ICT-FinD-Lab/Awesome-LLMs-for-AI-Research/blob/master/PDF/AI4AIR_Survey_v260601.pdf
š Project: https://ict-find-lab.github.io/Awesome-LLMs-for-AI-Research
ā GitHub: https://github.com/ICT-FinD-Lab/Awesome-LLMs-for-AI-Research
Stars & feedback warmly welcome š
Sharing our new survey, AI4AIR: A Comprehensive Survey on Large Language Models for AI Research š
Rather than treating LLMs as outside helpers for writing or literature review, we map how they enter the ML pipeline itself ā through a 2D taxonomy (research domain Ć pipeline stage) and 5 recurring roles:
š·ļø Annotator Ā· š§Ŗ Synthesizer Ā· āļø Optimizer Ā· š Evaluator Ā· šļø Orchestrator
Our key takeaway: how deeply an LLM can empower a stage is bounded by its validation cost ā which is exactly why data annotation & evaluation are already mature, while model design & full-loop orchestration remain hard.
š Paper (PDF): https://github.com/ICT-FinD-Lab/Awesome-LLMs-for-AI-Research/blob/master/PDF/AI4AIR_Survey_v260601.pdf
š Project: https://ict-find-lab.github.io/Awesome-LLMs-for-AI-Research
ā GitHub: https://github.com/ICT-FinD-Lab/Awesome-LLMs-for-AI-Research
Stars & feedback warmly welcome š
upvoted a paper 7 months ago
submitted a
paper to Daily Papers 7 months ago
liked a
dataset 9 months ago
upvoted a paper 10 months ago
upvoted 6
papers 11 months ago
WebSailor-V2: Bridging the Chasm to Proprietary Agents via Synthetic Data and Scalable Reinforcement Learning
Paper ā¢ 2509.13305 ā¢ Published ā¢ 93
Single-stream Policy Optimization
Paper ā¢ 2509.13232 ā¢ Published ā¢ 36
Why Language Models Hallucinate
Paper ā¢ 2509.04664 ā¢ Published ā¢ 200
Implicit Actor Critic Coupling via a Supervised Learning Framework for RLVR
Paper ā¢ 2509.02522 ā¢ Published ā¢ 26
Transition Models: Rethinking the Generative Learning Objective
Paper ā¢ 2509.04394 ā¢ Published ā¢ 29
Predicting the Order of Upcoming Tokens Improves Language Modeling
Paper ā¢ 2508.19228 ā¢ Published ā¢ 23
authored a
paper 12 months ago
upvoted 5
papers about 1 year ago
Improved Personalized Headline Generation via Denoising Fake Interests from Implicit Feedback
Paper ā¢ 2508.07178 ā¢ Published ā¢ 1
Seed Diffusion: A Large-Scale Diffusion Language Model with High-Speed Inference
Paper ā¢ 2508.02193 ā¢ Published ā¢ 139
NeuralOS: Towards Simulating Operating Systems via Neural Generative Models
Paper ā¢ 2507.08800 ā¢ Published ā¢ 81
WebSailor: Navigating Super-human Reasoning for Web Agent
Paper ā¢ 2507.02592 ā¢ Published ā¢ 127
LaMP-Cap: Personalized Figure Caption Generation With Multimodal Figure Profiles
Paper ā¢ 2506.06561 ā¢ Published ā¢ 2
upvoted a paper over 1 year ago
liked a
dataset over 1 year ago