2026-08-12 04:00 UTCOriginal source2 min readUpdated: 2026-08-12 08:41 UTC

When Chain-of-Thought Helps and When It Hurts: An Empirical Investigation of the Serial-Depth Bottleneck in LLM Reasoning

arXiv:2608.09942v1 Announce Type: new Abstract: It is widely assumed that chain-of-thought (CoT) prompting universally improves LLM reasoning. We investigate this through the conceptual framework of the H_dp bandwidth bound (Chen et al., 2024): although the formal bound binds only asymptotically (at astronomically large prompt lengths), it identifies a real architectural bottleneck -- serial computation exceeding a transformer's single-pass capacity must be externalised, which is what CoT does. Our central finding is a within-benchmark serial-depth gradient: single-pass (no-CoT) accuracy degrades monotonically with per-item serial depth, while CoT is approximately depth-invariant. We measure CoT effects across three instruction-tuned models (Qwen-2.5-7B/32B, Llama-3.1-8B) and five standard NLP benchmarks at practical context lengths. On high-depth P-complete tasks (GSM8K, MATH), CoT gives a +54 to +68 pp recovery gap across all models. On shallow TC^0 tasks (MMLU, ARC), CoT is structurally redundant (Delta in [0.0, +4.6] pp, no significant negative effect) -- though high no-CoT baselines (up to 95% on ARC) may reflect contamination, so this null is not a clean architectural test. The intermediate class L (HumanEval) shows a model-size-dependent transition: +23.2 pp (32B), +9.1 pp (8B), -28.7 pp (7B). The cross-benchmark depth-recovery correlation is Spearman rho = 0.661 (p = 0.007, n = 15); 9 of 15 benchmark-level McNemar tests are significant after Bonferroni correction. Pre-registered on OSF, our results indicate that CoT is not a universal reasoning enhancer but acts as a bandwidth bypass: it helps serial computation that strains single-pass capacity and is redundant for tasks that already fit.

SourcearXiv Computational LinguisticsAuthor: Tughanbulut Kurtulush

-->

[Submitted on 23 Jun 2026]

Title:When Chain-of-Thought Helps and When It Hurts: An Empirical Investigation of the Serial-Depth Bottleneck in LLM Reasoning

View a PDF of the paper titled When Chain-of-Thought Helps and When It Hurts: An Empirical Investigation of the Serial-Depth Bottleneck in LLM Reasoning, by Tughanbulut Kurtulush

View PDF HTML (experimental)

Abstract:It is widely assumed that chain-of-thought (CoT) prompting universally improves LLM reasoning. We investigate this through the conceptual framework of the H_dp bandwidth bound (Chen et al., 2024): although the formal bound binds only asymptotically (at astronomically large prompt lengths), it identifies a real architectural bottleneck -- serial computation exceeding a transformer's single-pass capacity must be externalised, which is what CoT does. Our central finding is a within-benchmark serial-depth gradient: single-pass (no-CoT) accuracy degrades monotonically with per-item serial depth, while CoT is approximately depth-invariant. We measure CoT effects across three instruction-tuned models (Qwen-2.5-7B/32B, Llama-3.1-8B) and five standard NLP benchmarks at practical context lengths. On high-depth P-complete tasks (GSM8K, MATH), CoT gives a +54 to +68 pp recovery gap across all models. On shallow TC^0 tasks (MMLU, ARC), CoT is structurally redundant (Delta in [0.0, +4.6] pp, no significant negative effect) -- though high no-CoT baselines (up to 95% on ARC) may reflect contamination, so this null is not a clean architectural test. The intermediate class L (HumanEval) shows a model-size-dependent transition: +23.2 pp (32B), +9.1 pp (8B), -28.7 pp (7B). The cross-benchmark depth-recovery correlation is Spearman rho = 0.661 (p = 0.007, n = 15); 9 of 15 benchmark-level McNemar tests are significant after Bonferroni correction. Pre-registered on OSF, our results indicate that CoT is not a universal reasoning enhancer but acts as a bandwidth bypass: it helps serial computation that strains single-pass capacity and is redundant for tasks that already fit.

Comments: 15 pages, 3 figures, 5 tables. Pre-registered study (OSF: this https URL). Data and code: this https URL

Subjects:

Computation and Language (cs.CL); Artificial Intelligence (cs.AI); Machine Learning (cs.LG)

Cite as: arXiv:2608.09942 [cs.CL]

(or arXiv:2608.09942v1 [cs.CL] for this version)

https://doi.org/10.48550/arXiv.2608.09942

arXiv-issued DOI via DataCite

Submission history

From: Tughanbulut Kurtulush [view email] [v1] Tue, 23 Jun 2026 12:38:52 UTC (538 KB)

Full-text links:

Access Paper:

View a PDF of the paper titled When Chain-of-Thought Helps and When It Hurts: An Empirical Investigation of the Serial-Depth Bottleneck in LLM Reasoning, by Tughanbulut Kurtulush

View PDF

HTML (experimental)

TeX Source

view license

Current browse context:

cs.CL

new | recent | 2026-08

Change to browse by:

cs cs.AI cs.LG

References & Citations

NASA ADS

Google Scholar

Semantic Scholar

Data provided by:

Bibliographic Tools

Bibliographic and Citation Tools

Bibliographic Explorer Toggle

Bibliographic Explorer (What is the Explorer?)

Connected Papers Toggle

Connected Papers (What is Connected Papers?)

Litmaps Toggle

Litmaps (What is Litmaps?)

scite.ai Toggle

scite Smart Citations (What are Smart Citations?)

Code, Data, Media

Code, Data and Media Associated with this Article

alphaXiv Toggle

alphaXiv (What is alphaXiv?)

Links to Code Toggle

CatalyzeX Code Finder for Papers (What is CatalyzeX?)

DagsHub Toggle

DagsHub (What is DagsHub?)

GotitPub Toggle

Gotit.pub (What is GotitPub?)

Huggingface Toggle

Hugging Face (What is Huggingface?)

ScienceCast Toggle

ScienceCast (What is ScienceCast?)

Demos

Replicate Toggle

Replicate (What is Replicate?)

Spaces Toggle

Hugging Face Spaces (What is Spaces?)

Spaces Toggle

TXYZ.AI (What is TXYZ.AI?)