The ProductConsistency dataset introduces 87k SFT samples and 869 RL samples to improve product identity preservation in image editing. It includes a benchmark for standardized evaluation and uses a cyclic consistency reward to enforce semantic product identity through caption similarity. Fine-tuning Qwen-Image-Edit-2511 and Flux.1-Kontext-dev shows a 5x reduction in character error rate and improved text rendering and visual quality.
arxiv
arXiv cs.AI
·
77d ago
·
research
ProductConsistency: Enhancing Product Identity in Image Editing
from English
Importance 2/3
New feature vs. leaders
arXiv cs.AI
Alibaba (Qwen)
Hugging Face
Evaluation & benchmarks
Image generation
Reasoning models
Related articles
arxiv
arXiv cs.LG
·
78d ago
·
4 views
NoiseTilt: Noise-Tilted Reverse Kernels for Diffusion Reward Alignment
NoiseTilt introduces NTRK, a reward-guided diffusion sampler that injects reward gradients via the noise term without altering the reverse kernel. By using a whitening operator, NTRK safely biases noise toward high reward, preserving sample quality while maintaining strong guidance. On aesthetic generation, NTRK achieves superior reward performance with 25 NFEs, reducing compute by 20× compared to state-of-the-art baselines.
lab
Hugging Face Blog
·
just now
Live
OpenEnv reproduces Surya Narreddi's watercolour painting model using TRL
An engineer has open-sourced a reproduction of Surya Narreddi’s viral project, which trains a coding model to paint watercolours using JavaScript and reinforcement learning. The implementation utilizes the TRL library and OpenEnv to create an end-to-end pipeline on Hugging Face for training, scoring, and inference.
arxiv
arXiv cs.CL
·
49d ago
·
14 views
GSM-Plus-BN introduces perturbation-based benchmark for Bangla math reasoning
The study introduces GSM-Plus-BN, a novel perturbed Bengali mathematical dataset derived from the English GSM-Plus benchmark and verified by human translators. This resource addresses the lack of systematic benchmarks for assessing model robustness in linguistically diverse regions like Bangladesh.
arxiv
arXiv cs.AI
·
71d ago
·
1 view
LLMs Benchmarked for Web Vulnerability Detection
A study evaluates six LLMs on detecting real-world web vulnerabilities in WordPress plugins, finding detection rates vary by model and prompt design. Claude Opus 4.6 achieved the highest detection rate at 63%, while Qwen 3.5 only reached 35%, and no model consistently identified all baseline vulnerabilities across iterations.
arxiv
arXiv cs.AI
·
71d ago
·
5 views
Benchmark Evaluation of Small Language Models for Arabic NLP
A benchmark of 240 Arabic test items across eight domains and ten skills assesses twelve small language models in zero-shot settings. Gemma 3 (12B) achieved the highest overall score (4.548/5), followed by Aya and C4AI Command Arabic, with performance linked more to Arabic alignment and instruction-following than model size. Common failure modes include prompt leakage, hallucination, and weak task adherence.