finding
active
finding:short-rationales-lora-cot-sometimes-improve-in-distribution-performance-but-do-not-reliably-reduce-cross-base-harmShort rationales (LoRA+CoT) sometimes improve in-distribution performance but do not reliably reduce cross-base harm
E2 finding showing CoT's limited benefit for OOD transfer, consistent with larger dr out of scope
Source paper
extracted_from(2025) · Edward Yi Chang · Kaya, Zeyneb N. · Ethan Chang
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.
- Interpretation of scope generalization results
- Figure 2 and Figure 8 illustrate RL-CAI at the Pareto frontier.
- Section 4.3 describes clamping at 40-60 led to better behavior than clamping at 20-80.
- Additional test-time computation from longer context is responsible for CoT's performance boost.hypothesis0.756Tested via filler tokens.
- CoT increases dr for OOD operands.
- Fine-tuning with chain-of-thought rationales aiming to reduce dr via procedural alignment.
- From filler token results.
- From decoupling of AOC and accuracy gain.