Dimensions overall score 8.0

GitHub Code Pulse

No public code linked for this paper yet.

• Mainstream PLM-based LJP methods lack explicit modeling of legal constituent elements and underlying causal logic, making them prone to learning spurious correlations and suffering from poor robustness.limitation 90%
• Existing causal LJP methods face inaccurate legal factor extraction with severe noise in real-world legal texts.limitation 85%
• Existing causal LJP methods face significant uncertainty in causal structure discovery due to Markov equivalence under sparse features.limitation 85%
• A coarse-to-fine hybrid extraction mechanism combining statistical sampling and LLM semantic reasoning can accurately identify and purify standard legal constituent elements.method 80%
• An LLM-assisted causal structure disambiguation mechanism using LLM as a constrained prior knowledge base can resolve structural uncertainty by conducting probabilistic evaluation and pruning on ambiguous causal directions.method 80%
• A causal-aware judgment prediction model constructed by explicitly constraining text attention intensity via generated causal graphs outperforms state-of-the-art baselines in predictive accuracy and robustness.result 75%
• The proposed method particularly improves performance in distinguishing confusing charges compared to existing approaches.result 70%
• Integrating LLM priors with statistical causal discovery creates an enhanced causal inference framework for LJP.technical 85%

Competitive landscape

Competitor map is still being generated for this paper. Enable generation or check back soon.

Startup potential card