Tag
This simulation study quantifies selection bias in retail inflation estimation and compares correction methods, finding that stratification generally outperforms inverse probability weighting in long-tail contexts with severe positivity violations.
This paper introduces a lightweight, offline-trained logit correction method for grammar constrained decoding that leverages internal parser and lexer states to restore the LM's true probability distribution without expensive online sampling, improving output quality while maintaining low inference latency.
OrDA is a framework that disentangles user access habits from genuine content interests in homepage marketing block recommendations using orthogonal regularization and causal intervention, achieving a 5.64% UCTR improvement on Zhima's homepage.
This paper introduces Retroactive Advantage Correction (RAC), a closed-form bias correction method for delay-aware RLHF that handles asynchronous reward signals by queuing and reinjecting delayed rewards with a V-trace-style clipped residual update.
This paper introduces SLC (State-space Logit Correction), which corrects per-item logit bias in knowledge tracing models using empirical-Bayes shrinkage via a Kalman smoother, improving AUC beyond global calibration techniques.
This paper introduces PRECISE, an extension of Prediction-Powered Inference that combines a small set of human labels with a large set of LLM judgments to produce unbiased and variance-reduced estimates of ranking evaluation metrics like Precision@K. The method is validated on the ESCI benchmark and in a production A/B test, where it correctly identified the best system variant using only 100 human labels, confirmed by a +407 bps sales improvement.
This paper identifies a bias in attention weights caused by quantizing keys in KV-cache compression for chunk-wise autoregressive video diffusion, and proposes a per-attention-score correction that removes the bias with negligible overhead, recovering near-BF16 video quality at INT2 quantization.
This paper presents PushCen-ADFL, a communication-efficient asynchronous decentralized federated learning framework that uses centroid-based messaging and bias-correction to improve accuracy and reduce communication overhead under heterogeneous conditions.