claim
active
claim:mere-expressivity-is-not-the-issue-the-question-is-learnability-under-gradient-optimization-from-a-data-mixtureMere expressivity is not the issue; the question is learnability under gradient optimization from a data mixture
All tasks are learnable by the models.
Source paper
extracted_from(2026) · Jing Huang · Daniel Wurgaft · Rachit Bansal · Laura Ruis +5
Neighborhood — ranked by edge-count
Papers (1)
paper
Claims (1)
claim
- Scale alone is not the mechanism enabling rare-task learning; multi-task methods can also enable itextendsClarification in related work.
Related by similarity (8)
cosine ≥ 0.65 · no typed edgeEntities in the same semantic neighborhood but without a typed relation to this one — candidates for new edges or unrecognized duplicates.
- Mechanistic explanation of why fine-tuning shifts persona vectors rather than directly learning narrow behaviors
- Contrast with reward model non-identifiability.
- Recommended strategy for gradient normalization.
- Parlog's merge process for client-server is unnecessarily complex; Linda's tuple operations remain flexible across problem variants.
- We hypothesize that degraded generalization on benchmarks like MMLU may reflect the computational demands of the tasks.hypothesis0.759Connecting the paper's task-difficulty findings to prior observations of weak generalization on complex QA benchmarks.
- Frequency alone is insufficient to predict learning order; both frequency and complexity are neededclaim0.758From complexity sweeps.
- Motivation for using sparsity-based dictionary learning on language models
- Selective pressure toward convergence via task generality