DreamReasoner-8B is an open-source block diffusion model that demonstrates strong long-chain-of-thought reasoning. A systematic study shows that small training block sizes preserve reasoning effectiveness, while large sizes degrade performance. Block-size curriculum learning gradually transitions training from fine to coarse blocks, enabling robust and generalizable reasoning across inference settings, with results competitive to Qwen3-8B on mathematical and code benchmarks.
arxiv
arXiv cs.CL
·
52d ago
·
research
DreamReasoner-8B: Block-Size Curriculum Learning for Diffusion Reasoning
from English
Importance 2/3
New feature vs. leaders
arXiv cs.CL
Alibaba (Qwen)
Mistral AI
OpenAI
Image generation
Open weights
Reasoning models
Related articles
arxiv
arXiv cs.CL
·
47d ago
·
6 views
Test-Time Steering Resolves Temporal Fact Conflicts in LLMs
Researchers identify parametric temporal conflicts in language models where outdated facts persist in parameters. They introduce Temporal Attractor Steering (TAS), a test-time method that resolves 29-57% of such conflicts without retraining, maintaining 85-99% accuracy on non-conflict queries and outperforming a baseline on three of four models.
arxiv
arXiv cs.CL
·
51d ago
·
1 view
AgentFinVQA: Auditable, On-Premise Financial Chart QA
AgentFinVQA introduces a multi-agent pipeline for financial chart question answering that ensures auditability and on-premise deployability without significant accuracy loss. It outperforms baseline models by +7.68 pp using a proprietary backbone and +4.84 pp with open-weights Qwen3.6-27B-FP8, while providing a confidence signal via verifier output that improves human review routing.
arxiv
arXiv cs.LG
·
52d ago
Reverse-Engineering Transformer Attention with Executable Programs
A new method uses program synthesis to generate Python programs that reproduce attention patterns in transformer models. These programs achieve over 75% average Intersection-over-Union similarity on held-out data and can replace up to 25% of attention heads with minimal impact on model performance, increasing perplexity by only 16% on average.
arxiv
arXiv cs.CL
·
52d ago
·
1 view
OmniAgent: Native Active Perception for Omni-Modal Understanding
OmniAgent introduces a POMDP-based iterative Observation-Thought-Action cycle for video understanding, enabling on-demand action execution to selectively distill audio-visual cues into persistent textual memory. It achieves state-of-the-art performance on ten benchmarks, with a 7B agent outperforming a 10× larger Qwen2.5-VL-72B model on LVBench (50.5% vs. 47.3%).
arxiv
arXiv cs.LG
·
53d ago
Baseline Evaluation of Open-Source LLMs for Multi-Label ATT&CK Classification
A ground-truth dataset of 2,076 human-annotated sentences from 83 complex CTI reports was constructed and mapped to 114 ATT&CK techniques with \k{appa} = 0.68 inter-annotator agreement. Seven open-source LLMs ranging from 8B to 236B parameters were evaluated, achieving a maximum micro-averaged F1 score of 0.22. Parameter size showed a statistically significant positive correlation with F1 score, while prompt strategy and temperature did not yield significant improvements, indicating current open-source LLMs are insufficient for production-grade ATT&CK classification.