Comprehensive Survey Maps LLM Reasoning Capabilities and Failure Modes Across 300+ Studies
Researchers have published a structured survey analyzing more than 300 recent papers on reasoning in Large Language Models (LLMs), introducing a taxonomy they call the 'Periodic Table of LLM Reasoning.' The survey covers paradigms ranging from chain-of-thought and mathematical reasoning to agentic and reinforcement learning-based reasoning, while cataloguing recurring failure modes. It aims to serve as a reference for building more robust and generalizable reasoning systems.
A preprint posted to arXiv on June 9, 2026 presents a comprehensive survey of LLM reasoning research, drawing on more than 300 papers from major academic databases including arXiv, Semantic Scholar, and the ACL Anthology. The authors introduce a structured taxonomy spanning nine major reasoning paradigms—including chain-of-thought, multi-hop, mathematical, commonsense, visual, temporal, code, retrieval-augmented, and reinforcement learning-based reasoning. Beyond cataloguing methods, the survey systematically analyzes prompting strategies, model architectures, training objectives, reward modeling, and evaluation benchmarks across these paradigms. A central finding is that despite notable progress, LLM reasoning remains inconsistent and sensitive to prompting strategies, task design, and model scale, with recurring failure modes including reasoning hallucinations, brittle multi-step inference, weak causal abstraction, and poor cross-domain generalization. The authors also identify emerging research directions such as meta-reasoning, self-evolving reasoning frameworks, multimodal reasoning, and socially grounded reasoning, positioning the survey as a roadmap for future work in the field.
What's missing
As a preprint, this survey has not yet undergone formal peer review, which means its taxonomy choices, paper selection criteria, and conclusions about failure modes have not been independently validated. The survey's scope is limited to papers available through the listed databases and may not capture all relevant work, particularly non-English or industry-internal research.
What different sources said
- arXiv cs.CLCenter
The Periodic Table of LLM Reasoning: A Structured Survey of Reasoning Paradigms, Methods, and Failure Modes
Related
Gut Bacteria Enzyme Found to Break Down Heat-Processed Food Compounds, Producing Novel Biogenic Amines
Researchers have discovered that an enzyme in common gut bacteria can degrade N-epsilon-carboxymethyllysine (CML), a compound formed during thermal food processing, producing previously unknown biogenic amines. The enzyme, ornithine decarboxylase SpeC from enterobacteria, acts on CML and related modified lysine derivatives through a low-level 'underground' catalytic activity. This finding suggests a previously unrecognized communication axis between thermally processed dietary compounds and gut microbial physiology, with potential implications for host health.
Full-Length Gene Sequencing Reveals Two Distinct Bacterial Communities in Black-Legged Ticks Expanding Into Canada
Researchers used Oxford Nanopore full-length 16S rRNA gene sequencing to characterize the microbiome of Ixodes scapularis black-legged ticks collected in Nova Scotia, Canada, distinguishing between tick-adapted bacteria and environmentally acquired bacteria. The study comes as I. scapularis — the primary vector of Lyme disease — is rapidly expanding northward into Canada due to climate change. The findings suggest that environmentally derived bacteria in tick microbiomes are not mere contamination, which has implications for how tick microbiome data is collected and interpreted across surveillance studies.
Study Identifies Metabolic Link Between Cell Envelope Stress and Biofilm Formation in Bacteria
Researchers have discovered that the metabolite acetyl-CoA directly inhibits enzymes that degrade the bacterial signaling molecule c-di-GMP, connecting cell envelope biosynthesis stress to biofilm formation in Pseudomonas aeruginosa. The study found that sub-inhibitory concentrations of antibiotics targeting early peptidoglycan biosynthesis — but not other antibiotic classes — elevate c-di-GMP levels by reducing phosphodiesterase activity, with acetyl-CoA competing for the enzyme active site. Because the relevant enzyme domain is broadly conserved across bacterial species, this checkpoint mechanism may be widespread and could have implications for understanding antibiotic-induced biofilm responses.