SPPO: Sequence-Level PPO for Long-Horizon Reasoning Tasks
Paper • 2604.08865 • Published • 12
AI & ML interests
Multilingual and multimodal LLM, data synthesis, complex reasoning with LLMs
Recent Activity
View all activity
Papers
SPPO: Sequence-Level PPO for Long-Horizon Reasoning Tasks
From Word to World: Can Large Language Models be Implicit Text-based World Models?
X1AOX1A
submitted a
paper to Daily Papers about 4 hours ago
X1AOX1A
authored a
paper 2 days ago
lllp11
updated a
dataset about 2 months ago
X1AOX1A
authored 10
papers 4 months ago
UniPoll: A Unified Social Media Poll Generation Framework via Multi-Objective Optimization
Paper • 2306.06851 • Published
LayAlign: Enhancing Multilingual Reasoning in Large Language Models via Layer-Wise Adaptive Fusion and Alignment Strategy
Paper • 2502.11405 • Published
VisCodex: Unified Multimodal Code Generation via Merging Vision and Coding Models
Paper • 2508.09945 • Published • 6
Enhancing Large Language Model Reasoning via Selective Critical Token Fine-Tuning
Paper • 2510.10974 • Published • 1
From Word to World: Can Large Language Models be Implicit Text-based World Models?
Paper • 2512.18832 • Published • 14
SeTAR: Out-of-Distribution Detection with Selective Low-Rank Approximation
Paper • 2406.12629 • Published
ImPart: Importance-Aware Delta-Sparsification for Improved Model Compression and Merging in LLMs
Paper • 2504.13237 • Published
G2: Guided Generation for Enhanced Output Diversity in LLMs
Paper • 2511.00432 • Published
MiLoRA: Harnessing Minor Singular Components for Parameter-Efficient LLM Finetuning
Paper • 2406.09044 • Published
PACIT: Unlocking the Power of Examples for Better In-Context Instruction Tuning
Paper • 2310.00901 • Published
eric18
updated a
model almost 2 years ago
eric18
updated a
model about 2 years ago