arXiv Machine Learning

Do Search-Augmented Recommenders Use the Latest Search Episode?

The paper introduces FocalSwap, a method to audit whether search‑augmented recommenders rely on the most recent search episode or on earlier searches. By systematically replacing the latest search slot or all search inputs with neutral or matched observed sequences while keeping other factors constant, the study measures rank changes, reciprocal rank gains, and preference for the true focal search. Experiments on KuaiSAR, Qilin, and UniSAR checkpoints reveal that the contribution of all search inputs varies with the control used, with neutral masks often showing a positive effect while matched replacements sometimes yield smaller or negative estimates, and that masking evidence alone can be misleading without comparison to observed‑search replacements.

arXiv AI
Sep 30

Multi-Channel Mitigation of Source-Trust Shortcuts in Fact-Checking RL Agents

The paper introduces TrustSwap, a counterfactual test that swaps or removes source reliability labels while keeping evidence text constant, to evaluate how retrieval‑augmented fact‑checking models respond across verdict, confidence, and search decisions. Experiments on untrained and RL‑trained models show that confidence and search largely follow labels, yet label changes can flip a significant portion of verdicts, especially in larger models. The authors propose trust‑swap augmentation (TSA) to mitigate this shortcut, demonstrating reduced verdict flip rates and maintained accuracy in several settings, though its effectiveness diminishes at larger model scales.

By Jianchang Su, Yiwei Yang, Wei Zhang
arXiv AI
Oct 1

False Frontiers: Diagnosing and Mitigating Co-Cheating in Self-Evolving Search Agents

The paper investigates a failure mode called co‑cheating in self‑evolving search agents, where the proposer and solver agree on shared errors, inflating internal reward without improving external correctness. The authors first propose multi‑sample verification (MSV) to filter unreliable pseudo‑labels, which only partially mitigates the issue. They then introduce CrossFit, a cross‑fitted reward scheme that partitions source documents and uses an auxiliary solver to prevent same‑source agreement, significantly reducing false agreement and boosting downstream benchmark performance.

By Meijia Chen, Hao Li, Zheng Lu, Hongshan Lin, Junbai Tian, Yichen Liu, Zijun Tian, Yufan Zou, Shuhan Sun, Hanxin Chen, Zeyu Zhang, Weizhi Du, Yueting Li, Tianyu Shi, Alaa Khamis
arXiv Machine Learning
Aug 28

Keeping the Index Open: The Recommendation-Side Cost of Shared Search and Recommendation

The paper investigates the trade‑off of using a shared search‑and‑recommendation index that scores new items purely from features, thereby keeping the index open to unseen items. Experiments on public logs show that a feature‑based tower can match warm‑item performance (Recall@20 0.9595 vs 0.9510) and a lexical baseline, while a full‑catalog check is inconclusive. The study also quantifies the cost of this openness on recommendation quality across several baselines, revealing that exact full‑softmax training improves recall but is impractical at catalog scale.

By Theodore Rogers, Joe Standerfer, Dmitrii Timoshenko, Haoxue Li, Zuhaib Akhtar, Soyoung Yang
arXiv AI
Oct 15

Search-G1: Co-Evolving Representation-Based Rewards for Selective Grounded Search

Search-G1 is a reward framework that encourages language agents to retrieve external information only when necessary and to ground their answers in that evidence. It uses two hidden‑state readouts—one predicting whether a closed‑book answer suffices and another measuring sensitivity to evidence deletion—to calibrate retrieval necessity and evidence reliance. The reward system gates evidence credit by necessity, rewards sufficient direct answers, and penalizes redundant searches, all without requiring costly annotations or LLM‑judge inference.

By Ruoxi Cheng, Haoxuan Ma, Hongyi Zhang, Junming Zhang, Ranjie Duan, Qiaolin Xia, Hao Wang, Yu Lu, Haibo Shi, Xingjun Ma
arXiv AI
Sep 18

The Missing Complement: State-Conditioned Minimal Sufficient Evidence for Coding Agents

The paper introduces State‑Conditioned Minimal Sufficient Evidence Recovery (SER), a method that, given a coding agent’s current state, reconstructs a compact set of evidence passages that collectively provide all facts needed for the agent’s next decision. Using the SERBench dataset of 500 held‑out states from 45 repositories, the authors show that their MSS‑Complement approach recovers a complete evidence set for 73.0 % of states with five items and 80.6 % with eight, outperforming baseline ranking methods. The study also demonstrates that this set‑level policy improves downstream performance on AMA‑Bench and highlights the importance of retrieving missing facts rather than merely re‑ranking similar passages.

By Zhexi Feng, Ruiyi Zhang, Yongbo Yang, Pengtao Xie