Research map / Architectures and efficiency / RNNs and normalization
LoRe: Adaptive Interaction-Evaluation Routing with Per-step Interaction Budgets for Iterative Graph Solvers
Jintao Li, Yong-Yi Wang, Zheng-an Wang, Heng Fan
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| use dynamic state-dependent re-routing instead of a static interaction support (LoRe-Static) (ablation) | LoRe on MIS conflict-descent relaxation · 40 ER/BA graphs, rho=0.08 | conflict convergence and legal solution size | Improved |
| combine a static skeleton with dynamic hotspots rather than purely dynamic greedy edge selection (ablation) | LoRe routing · 40 ER/BA graphs, rho=0.08 | stability of convergence | Improved |
| sweep proxy-score weight lambda_stab from 0 to 5.0 (ablation) | LoRe · DIFUSCO MIS n=5000 | quality retention: range 0.9 pp / 2.8 pp / 1.6 pp across densities | No clear effect |
| vary interaction budget rho in [0.05, 0.50] (ablation) | LoRe · MIS | quality retention: less than 3 pp variation | No clear effect |
| vary refresh interval R in [1, 50] (ablation) | LoRe · MIS | quality retention: less than 1.5 pp variation | No clear effect |
| apply LoRe per-step interaction budgeting to DIFUSCO MIS at larger graph size | DIFUSCO · MIS n=15k | peak memory (GB): 86.7 -> 7.32 | Improved |
| apply LoRe to DIFUSCO MIS beyond baseline OOM limit | DIFUSCO · MIS n=20k-50k | feasibility: baseline OOM; LoRe feasible up to n=50k | Improved |
| apply LoRe to DIFUSCO TSP at small n | DIFUSCO · TSP n=100 | end-to-end time (s): 0.34 -> 0.61 | Hurt |
| apply LoRe to DIFUSCO TSP at n=1000 | DIFUSCO · TSP n=1k | end-to-end time (s): 13.6 -> 0.94 | Improved |
| apply LoRe zero-shot to scale-free BA graphs with ER-pretrained model | DIFUSCO · MIS topology OOD | quality retention: 0.991 -> 0.756 (ER vs BA) | Hurt |
| apply LoRe to small MIS instances | DIFUSCO · MIS n=1k-3k | quality retention: 0.815, 0.804, 0.835 | Hurt |
| apply LoRe to COExpander MIS | COExpander · MIS V=2.4k | quality retention: 0.872 retention | Hurt |
Closest papers by meaning
- Get a GRIP, this will be a long TRIP: A Quantifiable Long-Range Framework for Verifying Over-squashing (NeurIPS 2026 · similarity 0.85)
- Benchmarking Structural Inference Methods for Interacting Dynamical Systems with Synthetic Data (NeurIPS 2024 · similarity 0.84)
- A Graph Dynamics Prior for Relational Inference (AAAI 2024 · similarity 0.84)
- Adaptive Message Passing: A General Framework to Mitigate Oversmoothing, Oversquashing, and Underreaching (ICML 2025 · similarity 0.84)
- Graph Mamba Operator: A Latent Simulator for Interacting Particle Systems (NeurIPS 2026 · similarity 0.84)
- DeFoG: Discrete Flow Matching for Graph Generation (ICML 2025 · similarity 0.83)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
