arXiv subjects

Amirhossein Zohrehvand

Publications and source records attributed to Amirhossein Zohrehvand.

4 recordsLinked to original sources

arXiv2026

Measuring Concept Content in Text from LLM Activations: ESG Evidence from Concept Vectors and Linear Probes

Recent work shows that Large Language Models (LLMs) encode more information in their activations than they show in their responses. We ask whether the information contained in activations could be used as a source of measurement for the presence of concepts. We compare two algorithms for extraction of concepts, i.e., Recursive Feature Machine (RFM) algorithm and linear probing against an embedding baseline, surface baselines, and the model's own response using a human-annotated Environmental, Social and Governance (ESG) dataset of financial texts. The best linear probe ranges from 0.6 to 3.3 percentage points below fine-tuned domain classifiers. The probes also outscore the model's own response in all comparisons. Furthermore, the probes outperform the RFM concept vectors in all accuracy comparisons. In a synthetic mixture test, their mean-pooled score tracks the number of concept sentences in a paragraph better than the concept-vector score. Put together, a linear probe on a frozen LLM is a cost-effective and generalizable alternative to fine-tuning a domain classifier.

cs.CL↗

arXiv2026

Loop-Back Authority in LLM Agent Teams: A Paired Experiment on Flat and Hierarchical Coordination

Does authority in AI teams improve the outcome? Organizational theory asserts that authority facilitates decision making, improving quality. Meanwhile, some nascent AI research suggests that revision under authority makes LLM output worse. Multi-agent LLM frameworks default to giving a Manager agent the authority to send a worker's output back for revision. Prior comparisons test the effect of authority using verifiable tasks. We conduct an experiment on an open-ended task, business-intelligence reporting, using a sample of 43 paired laptop products and 86 runs. Each report is written once by a hierarchical team and once by a flat team. We find that flat teams produce higher-quality reports, scoring higher on Utility (d = 0.42, p = 0.009) and Writing Clarity (d = 0.34, p = 0.030). The reports are the same length, but hierarchical team reports use 53% more hedging words such as "may" and "could", and each revision is associated with a 0.14-point drop in Writing Clarity on a 1 to 5 scale. Before any revision, the hierarchical team's first draft is indistinguishable from the flat team's report. In other words, the quality gap can be traced to revision. Authority improves quality when the Manager can verify the work, else when it can only provide feedback it has a negative effect on quality.

cs.MA↗

arXiv2026

Recipes for Creativity: Iterative Generation and Evaluation in Large Language Models

Generative models are often evaluated through singular artifacts, whereas human creativity typically emerges through iterative generation, appraisal, and refinement. This pilot study examines whether iterative search improves LLM creativity by adapting FunSearch to recipe generation for the 2024 Pillsbury Bake-Off and evaluating outputs against human benchmarks using TTCT-based LLM evaluation. Across two experiments, we test iteration count, generator temperature, and in-loop selection-scorer model size. Results show that iterative generation-selection can produce recipes with creativity scores comparable to human benchmarks, but additional iterations alone do not improve creativity. The in-loop evaluator matters most: a smaller selection scorer yields significantly higher scores across most TTCT dimensions, while temperature has limited effects except for originality. These findings suggest that evaluator design is a first-order design variable in subjective creative search.

cs.AI↗

arXiv2024

The Impacts of AI Avatar Appearance and Disclosure on User Motivation

This study examines the influence of perceived AI features on user motivation in virtual interactions. AI avatars, being disclosed as being an AI, or embodying specific genders, could be used in user-AI interactions. Leveraging insights from AI and avatar research, we explore how AI disclosure and gender affect user motivation. We conducted a game-based experiment involving over 72,500 participants who solved search problems alone or with an AI companion. Different groups experienced varying AI appearances and disclosures. We measured play intensity. Results revealed that the presence of another avatar led to less intense play compared to solo play. Disclosure of the avatar as AI heightened effort intensity compared to non-disclosed AI companions. Additionally, a masculine AI appearance reduced effort intensity.

cs.HC↗