Causal evaluation framework for learnability of formal language tasks in LMs
A new arXiv preprint proposes a causal framework for evaluating how much task-specific data language models need to learn a given task. The authors use formal languages generated by probabilistic finite automata as a controlled testbed, introducing the 'binning semiring' algebraic object to control property frequency in training corpora. Experiments show that standard correlational evaluation practices produce incorrect learnability conclusions due to confounders, with implications for how natural-language task learning is studied.
Related guides (1)
Related events (8)
Study compares human and LLM active causal reasoning, finding LLMs less efficient but near human-level on conjunctive rules
A new arXiv paper investigates whether active exploration reduces the 'conjunctive handicap' in causal learning, using a blicket detector task with adult participants who could freely intervene to identify causal objects. Results show active exploration substantially improves human conjunctive causal reasoning, though conjunctive rules still require more tests than disjunctive ones. State-of-the-art LLMs approach human-level hypothesis inference accuracy but show less efficient exploration strategies and similar conjunctive-disjunctive performance gaps, raising questions about the nature of LLM causal reasoning.
CausaLab: Scalable Benchmark for Interactive Causal Discovery by LLM Agents
CausaLab is a new evaluation environment that tests LLM agents on interactive causal discovery tasks, requiring them to recover both causal graphs and structural equations from synthetic laboratory episodes governed by randomly sampled structural causal models (SCMs). The benchmark separates predictive accuracy from genuine causal understanding, revealing a persistent gap: GPT-5.2-high achieves 92% task accuracy in a 6-node observational setting but only 0.471 all-edge F1 for mechanism recovery. Mixed observation-intervention strategies improve structural fidelity, while pure intervention strategies underperform on both metrics. Premature stopping is identified as a key agent weakness, partially mitigated by prompting models to verify hypothesis-data consistency.
Language Models Learn Constructional Semantics, Not To Mention Syntax: Investigating LM Understanding of Paired-Focus Constructions
This paper investigates whether language models can learn the semantics of rare English constructions (e.g., 'let alone', 'much less'), constructing a novel dataset to test form-meaning pairing understanding. Testing models across parameter counts, architectures, and pretraining dataset sizes, the authors find that modestly sized open-source models can grasp Paired-Focus construction semantics, while models trained on human-scale data fail. Training dynamics analysis reveals that semantic understanding of these constructions emerges later than syntactic knowledge and correlates with gains in world knowledge more broadly.
Study finds shared pattern-matching mechanisms underlie both human and LLM everyday reasoning errors
A new arXiv paper evaluates human participants and 25 LLMs on commonsense causal reasoning tasks, finding similar error patterns in both groups. The authors identify specific attention heads driving LLM responses that implement pattern-matching, and show these heads can predict human reasoning errors caused by superficially irrelevant prompt details. The findings challenge the common assumption that human reasoning relies on principled abstract world models while LLMs merely pattern-match, suggesting both may share a more unified cognitive mechanism.
The Abstraction Gap in Vision-Language Causal Reasoning
Researchers introduce a dual-probe methodology and the CAGE benchmark (49,500 questions across 5,500 images) to distinguish linguistic plausibility from faithful causal reasoning in vision-language models. An Abstraction Gap (AG) metric quantifies the normalized performance difference between text-only and chain-of-reasoning probes. Evaluating eight VLMs, seven exhibit AG exceeding 0.50—generating fluent causal text but failing structured causal chain tasks—while one model achieves near-zero AG, suggesting architectural and pretraining choices are decisive. Fine-tuning on 45,000 chain-annotated examples fails to close the gap, pointing to a fundamental capability distinction.
Revisiting LLM systematicity in negation understanding via in-context learning
A new arXiv preprint analyzes how well large language models handle negation from two angles: behavioral systematicity (whether models correctly recognize negation expressions and scope) and representational systematicity (whether function vectors can be reliably constructed from in-context examples). Results show LLMs partially succeed at negation cue recognition via in-context learning but struggle with scope recognition, with performance varying by output format. Function vectors can be composed for cue extraction but are harder to extract for scope recognition tasks.
Tracing the Emergence of Human-Like Pragmatic Reasoning in LLMs Across Languages
Researchers conducted a population-matching experiment evaluating 25 LLMs on conditional inference tasks across four languages, comparing model behavior to matched human populations. The study finds that LLMs function as accurate semantic operators but systematically fail to capture pragmatic enrichments—context-sensitive inferences beyond literal logical meaning—that humans apply effortlessly. Model performance on pragmatic reasoning is not predicted by open vs. closed weights, training orientation, or architecture type, suggesting pragmatic reasoning remains an emergent and unreliable capability. The findings contribute to ongoing debates about whether LLMs reason like humans or merely approximate surface-level linguistic patterns.
Semi-supervised framework scales LLM reasoning with minimal labeled data via lightweight verifier
A new arXiv preprint proposes a semi-supervised framework for training LLMs to reason with very few labeled examples, using a lightweight classifier to judge the validity of intermediate reasoning traces. An entropy-based confidence threshold filters unreliable pseudo-labels before fine-tuning. Experiments on math reasoning (Orca-Math subset) and visual QA (GQA) show accuracy comparable to using 10-15x more labeled data. The approach reduces dependence on expensive answer-level supervision by turning verification into a data-creation mechanism.