Unverified50% confidenceBenchmarkExact time
Compared with hard filtering (directly discarding conflicting pairs) and recent constrained-optimization and multi-objective optimization baselines, CoDPO achieves a better trade-off between drift and preference; hard filtering loses the preference signal in the data, whereas CoDPO's downweighting strategy retains this information and only weakens its adversarial effect.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
11/6/2026
First Seen
Valid until: 2/4/2027
Sources
Related Entities
Related Claims
UnverifiedOn four main comparison domains, CoDPO consistently reduced constraint drift while largely preserving the original DPO's preference learning ability.76% similarUnverifiedOver-filtering leads to lower recall; when filter conditions are too narrow and the candidate set is empty, the system needs to degrade gracefully rather than return irrelevant snippets70% similarUnverifiedReduced variance leads to less clipping, because in rejection sampling an overly volatile estimate of the acceptance probability requires more aggressive clipping to ensure numerical stability, and clipping introduces additional distributional bias.70% similarUnverifiedNegative constraints narrow the probability space by explicitly excluding undesired outputs, are equally important as positive instructions, and can reduce deviations caused by the model's creative improvisation.70% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1141630API
curl https://kongchang.com/api/v1/knowledge/claims/1141630MCP
get_claim(id=1141630)