arXiv:2606. 28471v1 Announce Type: new Abstract: Model capability is the central variable in LLM pre-training, yet is never observed directly: data shapes it prospectively, while evaluation reveals it only retrospectively, compressing samples, prompts, decoding, and scoring rules into one noisy score.
By Zhixuan Li, Jiangan Yuan, Han Xu
The paper investigates how self‑improving large language models (LLMs) can suffer from the winner’s curse when selecting better candidates based on noisy evaluation sets. By modeling correlated errors and empirically testing Qwen models that rewrite their own instructions, the authors find that most proposals after the first are harmful and that the reported gains on the selection set often overstate true held‑out performance. The study shows that larger selection sets can improve held‑out gains on some tasks (e.g., TREC) but not others (e.g., GSM8K), and that common acceptance rules do not outperform simple greedy selection across entire runs.
By Litao Hu, Yutong Tang
The study examines how two small instruction‑tuned language models, Qwen2.5‑1.5B and Llama‑3.2‑1B, respond to user pushback on TriviaQA. When initially correct, the models flip to a wrong answer in about 42–43% of cases, with the effectiveness of different pushback styles varying by model. Attempts to decode capitulation from the pre‑response residual stream fail under a rigorous validation protocol, revealing overfitting and a measurement hazard that underestimates capitulation by 18–24 percentage points.
By Saad Aamir, Muhammad Awais Bin Adil
The paper evaluates how well typed decision models, which output probability distributions, actually provide reliable guarantees. By comparing open-weight Laya and Jev 1.13 across 190 evaluations, the study finds that point‑estimate thresholds frequently break promised accuracy, while a Beta lower‑bound rule and Learn‑then‑Test approach maintain guarantees more consistently. The work also explores calibration across languages, the impact of noise, data requirements for fine‑tuning, and the limitations of confidence in predicting outcomes or providing faithful explanations.
By Pedro Carvalho Brom
arXiv:2609.38699v1 Announce Type: new
Abstract: A cumulative token cap can fall inside a mathematical derivation, forcing a test-time controller to choose between stopping at the cap (strict) and all...
By Guilin Zhang, Ziqi Tan, Wulan Guo, Kai Zhao, Hongyun Yang, Mei Luo, Qi Ning, Feng Yang
The paper investigates whether large language models’ reasoning traces truly contain early, informative signals or merely reflect budget and difficulty confounds. Using a restart‑controlled truncation probe, the authors compare continuation success rates against from‑scratch restart curves across 178 problem‑model pairs, finding that only one case shows prefix‑limited success and that continuing a model’s own prefix generally outperforms restarting. A difficulty‑controlled test and two generation‑free analyses reveal that early internal signals do not carry outcome information beyond a problem‑difficulty baseline, underscoring the need for proper counterfactual controls.
By Yigit Utku Bulut