Computer Science > Computation and Language
Title:Scaling Laws for Deadline Proximity in Instruction-Tuned Language Models
Abstract:Scaling laws describe how language model error falls with parameters and data, and more recently with inference-time compute. We study a further variable that costs nothing to change: how soon the user says the answer is needed. We add a one-sentence deadline to the system prompt, ranging from six months to fifteen minutes, and evaluate ten instruction-tuned models from two families (Qwen2.5, 1.5B to 72B parameters; Llama 3.x, 1B to 70B) on GSM8K, MATH-500 and HumanEval, with five seeds and a time-matched control sentence. Between six months and one hour, GSM8K error follows with , so a closer deadline acts as a multiplier on effective parameter count. The fitted exponent is (95% CI 0.086 to 0.117) for Qwen2.5 and 0.089 (0.072 to 0.105) for Llama. Each halving of the time to deadline is therefore worth about 7% more parameters, and a one-hour deadline is worth a factor of 2.3. A law fitted without the 72B model predicts that model's errors to within 0.2 points. Below one hour the relation breaks down. At fifteen minutes error rises again and responses are 43% shorter than under no deadline. When the deadline is described as already missed, error exceeds the no-deadline control in all ten models, and 31% of Qwen2.5-7B responses open with an apology. We discuss habituation over long conversations and what would happen to the effect if deadline conditioning were widely used.
Submission history
From: Hanna Ekström[v1] Sun, 1 Nov 2026 08:39:22 UTC (214 KB)