Research map / Learning theory and algorithms / Sample complexity
Teaching via Best-Case Counterexamples in the Learning-with-Equivalence-Queries Paradigm
Akash Kumar, Yuxin Chen, Adish Singla
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 |
|---|---|---|---|
| best-case counterexamples instead of random counterexamples | teacher in LwEQ · Axes-aligned hyperplanes | query complexity: Omega(log d) -> 1 | Improved |
| best-case counterexamples instead of worst-case counterexamples | teacher in LwEQ · Axes-aligned hyperplanes | query complexity: Omega(d + log(n + 1)) -> 1 | Improved |
| best-case counterexamples instead of random counterexamples | teacher in LwEQ · Monotone monomials | query complexity: Omega(n) -> 1 | Improved |
| best-case counterexamples instead of worst-case counterexamples | teacher in LwEQ · Orthogonal rectangles | query complexity: Omega(d · log(n + 1)) -> 2 | Improved |
| best-case counterexamples instead of random counterexamples | teacher in LwEQ · Orthogonal rectangles | query complexity: Omega(d) -> 2 | Improved |
| binary yes/no feedback instead of best-case counterexamples | teacher in LwEQ · Monotone monomials | query complexity: 1 -> Omega(2n) | Hurt |
Closest papers by meaning
- The Signal is in the Steps: Local Scoring for Reasoning Data Selection (ICML 2026 · similarity 0.85)
- Few-Shot Knowledge Distillation of LLMs With Counterfactual Explanations (NeurIPS 2025 · similarity 0.85)
- DART-Math: Difficulty-Aware Rejection Tuning for Mathematical Problem-Solving (NeurIPS 2024 · similarity 0.84)
- On the Complexity of Teaching a Family of Linear Behavior Cloning Learners (NeurIPS 2024 · similarity 0.84)
- Distilling What Matters: Confidence-Aware Selective Distillation for Large Language Models (NeurIPS 2026 · similarity 0.84)
- Learning to Solve Complex Problems via Dataset Decomposition (NeurIPS 2025 · similarity 0.84)
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.
