arXiv:2609.06473v1 Announce Type: new Abstract: Inference-time activation steering enables behavioral control of large language models without parameter modification, while post-training quantization...
The study investigates how post‑training quantization (PTQ) affects proactive interference (PI) in large language models. Using bitsandbytes, the authors compare FP16, INT8, and INT4/NF4 precision across three instruction‑tuned models and find that INT4 quantization markedly degrades accuracy under high interference, with INT8 also incurring a smaller penalty in two of the three models. The degradation is linked to increased same‑key intrusion errors and originates in the quantized transformer backbone rather than the output layer.
Standard decoding rules for autoregressive language models promote diversity by rescaling the full next-token distribution or truncating its low-probability tail. These strategies overlook a common re...
The study investigates how quantization affects subliminal preferences in AI models, specifically a hidden "favourite animal: panda" preference implanted in Qwen2.5-7B-Instruct. While 8‑bit quantization preserves the preference, 4‑bit quantization produces divergent effects: Half‑Quadratic Quantization amplifies the preference to 64% panda probability, whereas NormalFloat suppresses it to 17%. The results show that the arrangement of rounding errors, not just their magnitude, drives these shifts, indicating that a 16‑bit audit cannot reliably predict real‑world behaviour under deployment‑level compression.
The paper investigates where post‑training quantization (PTQ) harms large language models (LLMs) and how to best allocate a limited precision budget. By causally raising each layer to 8‑bit precision across nine open‑weight models, the authors find that quantization damage is diffuse rather than concentrated in specific task circuits or weight statistics, and that globally refining quantization granularity outperforms selectively protecting the most recoverable layers. They also observe that the residual accuracy loss is budget‑limited and that peak recovery locations correlate with architecture within families but not across families.
XTC (Exclude Top Choices) is a lightweight, head‑aware decoding operator that improves diversity in autoregressive language models by removing overly probable tokens that dominate the next‑token distribution. It works by identifying tokens above a plausibility threshold, probabilistically excluding the dominant choices, and renormalizing the remaining distribution. Across 60 experiments on models such as Gemma 3 and DeepSeek R1, XTC boosts Distinct‑2 scores by 11–15 % and cuts repeat trigrams by 27–47 %, while a Mechanical Turk study shows a 62.3 % preference for XTC‑generated text without loss of fluency.
arXiv:2609.37887v1 Announce Type: new Abstract: Activation and key-value cache precision change what a quantized language model computes without altering its stored weights. Direct weight-code bounds...
The study investigates how post‑training quantization (8‑bit vs. 4‑bit) affects language‑model agents’ ability to recover from temporary tool failures. Using Llama‑3.1‑8B‑Instruct and Qwen2.5‑7B‑Instruct on twenty deterministic tool‑use tasks across five prompts, the authors find that recovery performance varies with prompt, evaluation target, and scoring policy, sometimes favoring 8‑bit, sometimes 4‑bit, and even reversing direction under different conditions. The results highlight that a single prompt, task set, or scoring rule cannot reliably determine the robustness of quantized agents. whyItMatters":"The findings emphasize that evaluation design choices critically influence conclusions about quantized‑agent robustness, underscoring the need for comprehensive, matched‑task assessments before deployment decisions are made."
arXiv:2608.28809v1 Announce Type: new Abstract: Extreme low-bit inference offers a route toward smaller models and constrained deployment. Ternary language models restrict weights to $\{-1,0,+1\}$, a...
Post-training quantization reduces the cost of deploying language-model agents, but its effect on recovery from temporary tool failures can depend on how recovery is evaluated. We compare 8-bit and 4-...
This study independently reproduces the dissociation reported by Zhao (2026) regarding chain-of-thought entropy in large language models. It confirms that the shape of the entropy trajectory predicts answer correctness, while the total entropy drop magnitude does not, across four open-weight models and two benchmarks (GSM8K and MATH‑500). The reproduction also maps settings where the magnitude signal holds or fails and documents protocol differences not reported in the original work.
The paper evaluates nine on‑device named‑entity recognition models ranging from classical taggers to large language models, measuring not only accuracy but also latency and output validity. Using a silver‑gold benchmark derived from an LLM judge panel and a human‑validated corpus, the study shows that encoder‑based models achieve comparable accuracy to a 4 B instruct LLM while being much smaller, faster, and producing no malformed output. Confidence calibration of GLiNER is analyzed, revealing over‑confidence but improved reliability after temperature scaling and thresholding.