Research map / Large language models / Language, safety and interpretability
TRENDNERT: A Benchmark for Trend and Downtrend Detection in a Scientific Domain
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 full text instead of abstracts (ablation) | trend detection baseline · TRENDNERT | MAP: .36 -> .07 | Hurt |
| use full text instead of abstracts (ablation) | trend detection baseline · TRENDNERT | R@50 avg: .50 -> .25 | Hurt |
| remove stratification (random sampling) (ablation) | trend detection baseline · TRENDNERT | MAP: .36 -> .30 | Hurt |
| remove stratification (ablation) | trend detection baseline · TRENDNERT | R@50 avg: .50 -> .50 | No clear effect |
| remove stratification (ablation) | trend detection baseline · TRENDNERT | R@50 max: .61 -> .53 | Hurt |
| use 16 year interval instead of 8 (ablation) | trend detection baseline · TRENDNERT | MAP: .36 -> .32 | Hurt |
| use 16 year interval instead of 8 (ablation) | trend detection baseline · TRENDNERT | R@50 avg: .50 -> .46 | Hurt |
| use 16 year interval instead of 8 (ablation) | trend detection baseline · TRENDNERT | R@50 max: .61 -> .61 | No clear effect |
| use k-means instead of GMM (ablation) | trend detection baseline · TRENDNERT | MAP: .36 -> .34 | No clear effect |
| use k-means instead of GMM (ablation) | trend detection baseline · TRENDNERT | R@50 avg: .50 -> .53 | Mixed |
Closest papers by meaning
- Future Language Modeling from Temporal Document History (ICLR 2024 · similarity 0.83)
- DATE-LM: Benchmarking Data Attribution Evaluation for Large Language Models (NeurIPS 2025 · similarity 0.82)
- A decoder-only foundation model for time-series forecasting (ICML 2024 · similarity 0.81)
- Training-free LLM-generated Text Detection by Mining Token Probability Sequences (ICLR 2025 · similarity 0.81)
- CruxBench: A Benchmark of Information Discovery (NeurIPS 2026 · similarity 0.81)
- Arrows of Time for Large Language Models (ICML 2024 · similarity 0.81)
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.