[
    {
        "id": "osp-22802",
        "type": "article-journal",
        "title": "Understanding Confabulation and Rethinking Reconstruction in Activation Explanations",
        "author": [
            {
                "family": "Lek",
                "given": "Gert"
            },
            {
                "family": "Xia",
                "given": "Zixuan"
            },
            {
                "family": "Chen",
                "given": "Pin-Yu"
            },
            {
                "family": "Chen",
                "given": "Lydia Y."
            }
        ],
        "URL": "https://omanscience.com/en/articles/understanding-confabulation-and-rethinking-reconstruction-in-activation-explanations",
        "language": "en",
        "issued": {
            "date-parts": [
                [
                    2026
                ]
            ]
        },
        "abstract": "Natural Language Autoencoders (NLAs) produce unsupervised text explanations of a model's activations: a verbalizer describes an activation and a reconstructor learns to recover it from this text. Under the established point-reconstruction NLA training recipe, explanations become more useful for predicting model behavior while also increasingly introducing unsupported details and exhibiting writing defects. To assess these changes separately, we introduce a standardized evaluation framework for unstructured NLA explanations, measuring information recoverable from explanations, contextual support for their claims, and writing quality. To address confabulation and writing defects, we move beyond predicting a single activation: explanations can distinguish distributions of possible activations even when their means and optimal point-reconstruction rewards are identical. We introduce Flow-NLA, which models the distribution of activations compatible with an explanation and trains the verbalizer using a diffusion likelihood bound. Across Qwen, Gemma, and Apertus, this richer signal retains the utility gains of point reconstruction while curbing the growth of confabulation and writing defects, opening up a direction for improving activation-derived training to encourage more informative, supported, and readable explanations. Code and evaluation prompts will be made publicly available upon acceptance."
    }
]