Test-Time Scaling via Error Localization

• URL: http://arxiv.org/abs/2607.21453v2
• Date: Fri, 24 Jul 2026 02:14:44 GMT
• Title: Test-Time Scaling via Error Localization
• Abstract summary: Scaling inference-time computation has emerged as a reliable method to improve the performance of large language models.<n>We introduce Test-Time Scaling via Error Localization (TTEL), an inference-time algorithm that utilizes fixed or environment feedback to perform token-level error localization.
• Score: 15.13459650752427
• License: http://creativecommons.org/licenses/by-nc-nd/4.0/
• Abstract: Scaling inference-time computation has emerged as a reliable method to improve the performance of large language models on complex reasoning and programming tasks. However, standard approaches such as independent sampling and sequential multi-turn refinement operate without token-level credit assignment, resulting in computational inefficiency, since valid reasoning prefixes are frequently discarded. In this work, we introduce Test-Time Scaling via Error Localization (TTEL), an inference-time algorithm that utilizes fixed or environment feedback to perform token-level error localization. By comparing conditional probabilities under informed feedback against a null-context baseline, TTEL isolates the step at which an error occurred. The algorithm then truncates the trajectory and branches a new generation, maximally reusing the valid prefix. Extensive evaluations demonstrate that TTEL establishes strictly dominating Pareto frontiers across sequential reasoning domains, measured by pass-at-k vs. generated-token cost. With Qwen3-8B on LiveCodeBench, TTEL attains a pass@64 of 71.0% while generating approximately half as many tokens as independent sampling (360.4k vs. 735.0k). Generalizing to math benchmarks AIME-2025 and HMMT-2025, TTEL cleanly outperforms competing test-time baselines across both Qwen3-8B and Qwen3-4B-Thinking-2507.

Related papers

• Thought-Level Beam Search for Reasoning [39.76902449987798]
  Test-time compute scaling is a primary driver of performance in large reasoning models.<n>We formalize test-time reasoning as a constrained compute allocation problem over partial trajectories.<n>We introduce Gambit, an inference algorithm that executes emphthought-level beam search
  arXiv  Detail & Related papers  (2026-08-08T09:00:01Z)
• Constrained Decoding for Diffusion Language Models via Efficient Inference over Finite Automata [57.27430779838529]
  We present an exact and tractable algorithm for sampling from the constrained mean-field posterior under any constraint expressible as a finite automaton.<n>The approach guarantees constraint satisfaction by construction, supports both greedy and sampling-based decoding, and is compatible with parallel and block-wise decoding.<n> Empirical evaluations on Dream-7B and LLaDA-8 show substantial accuracy gains across various tasks.
  arXiv  Detail & Related papers  (2026-07-08T05:48:57Z)
• Efficient and Trainable Language Model Test-Time Scaling via Local Branch Routing [65.59464371673623]
  Local Branch Routing (LBR) is a token-level test-time scaling framework.<n>It expands a small local lookahead tree, forwards all sampled branches through the language model, and uses a lightweight router to select the depth-1 subtree to commit.<n>LBR allows each token decision to use evidence beyond the root next-token distribution while avoiding full solution-level search.
  arXiv  Detail & Related papers  (2026-06-24T03:42:44Z)
• Closing the Loop on Latent Reasoning via Test-Time Reconstruction [45.08180971427891]
  Recent work moves intermediate reasoning from natural-language traces into latent or cache-level representations.<n>We propose ReLAT, a self-supervised test-time training method that closes this loop using the query itself as the reference.<n>We show that ReLAT consistently improves over single-model inference, text-based collaboration, open-loop latent collaboration, and alternative test-time training objectives.
  arXiv  Detail & Related papers  (2026-06-04T14:54:40Z)
• LaTER: Efficient Test-Time Reasoning via Latent Exploration and Explicit Verification [35.08680804423239]
  Chain-of-thought (CoT) reasoning improves large language models (LLMs) on difficult tasks, but it also makes inference expensive.<n>We propose Latent-Then-Explicit Reasoning (LaTER), a two-stage paradigm that first performs bounded exploration in a continuous latent space.<n>LaTER projects final-layer hidden states back to the input embedding space, preserves the latent KV cache, and uses entropy and model-native stop-token probes to decide when to switch.
  arXiv  Detail & Related papers  (2026-05-08T06:23:58Z)
• Algorithm-Based Pipeline for Reliable and Intent-Preserving Code Translation with LLMs [3.4257278503723576]
  Direct one-shot translation often fails to preserve program intent, leading to errors in control flow, type handling, and I/O behavior.<n>We propose an algorithm-based pipeline that introduces a language-neutral intermediate specification to capture these details before code generation.
  arXiv  Detail & Related papers  (2026-02-18T00:34:29Z)
• Broken Tokens? Your Language Model can Secretly Handle Non-Canonical Tokenizations [83.93566096400723]
  We find that instruction-tuned models retain up to 93.4% of their original performance when given a randomly sampled tokenization.<n>Character-level segmentation improves string manipulation and code understanding tasks by up to +14%.<n>Right-aligned digit grouping enhances large-number arithmetic by +33%.
  arXiv  Detail & Related papers  (2025-06-23T18:02:26Z)
• Accelerated Test-Time Scaling with Model-Free Speculative Sampling [58.69141724095398]
  We introduce STAND (STochastic Adaptive N-gram Drafting), a novel model-free speculative decoding approach.<n>We show that STAND reduces inference latency by 60-65% compared to standard autoregressive decoding.<n>As a model-free approach, STAND can be applied to any existing language model without additional training.
  arXiv  Detail & Related papers  (2025-06-05T07:31:18Z)
• Learning Adaptive Parallel Reasoning with Language Models [70.1745752819628]
  We propose Adaptive Parallel Reasoning (APR), a novel reasoning framework that enables language models to orchestrate both serialized and parallel computations end-to-end.<n> APR generalizes existing reasoning methods by enabling adaptive multi-threaded inference using spawn() and join() operations.<n>A key innovation is our end-to-end reinforcement learning strategy, optimizing both parent and child inference threads to enhance task success rate without requiring predefined reasoning structures.
  arXiv  Detail & Related papers  (2025-04-21T22:29:02Z)
• Fast Controlled Generation from Language Models with Adaptive Weighted Rejection Sampling [90.86991492288487]
  evaluating constraint on every token can be prohibitively expensive.<n> LCD can distort the global distribution over strings, sampling tokens based only on local information.<n>We show that our approach is superior to state-of-the-art baselines.
  arXiv  Detail & Related papers  (2025-04-07T18:30:18Z)
• LTM: Scalable and Black-box Similarity-based Test Suite Minimization based on Language Models [0.6562256987706128]
  Test suites tend to grow when software evolves, making it often infeasible to execute all test cases with the allocated testing budgets. Test suite minimization (TSM) is employed to improve the efficiency of software testing by removing redundant test cases. We propose LTM (Language model-based Test suite Minimization), a novel, scalable, and black-box similarity-based TSM approach.
  arXiv  Detail & Related papers  (2023-04-03T22:16:52Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.