arXiv subjects

Jiahui Ge

Publications and source records attributed to Jiahui Ge.

5 recordsLinked to original sources

arXiv2026

What Does a Repair Agent Gain from Remembering Its Failure? An Exploratory Study on Real Python Bugs

When a coding agent fails a repair task, it is often retried with a record of the failure: the transcript, a self-written lesson, or extracted test signals. We ask what such failure memory changes in the retry. On 75 BugsInPy bugs, we retried each of 181 failed attempts of gpt-5.4-mini under five conditions, with a predeclared analysis whose deviations we report. Two predeclared follow-ups repeated the plain-retry, transcript and lesson conditions with the larger gpt-5.4 and added a notice-only control. On the bug-revealing tests, every memory condition gained over a plain retry in point estimate (+5.5 to +9.9 points over 29.8%), but none significantly. Five further findings stand out. First, a model-written lesson showed no advantage over the transcript it summarises (-1.7 points, 95% CI [-8.7, +4.9]; -3.5 for gpt-5.4), although it costs an extra model call. Second, the transcript may do more than tell the agent that it failed: a notice alone produced no significant change, while the transcript resolved 12.2 points more failures than the notice for gpt-5.4-mini (CI [+4.2, +20.0], partial run of 156 failures; the only significant one of twelve predeclared contrasts, and not significant for gpt-5.4). Third, post hoc, fewer retries with memory ended without modifying any Python file, while the share of editing retries that touched a file the fix changes was near ceiling. Fourth, in a post hoc full-test-suite check, retries that had also seen the previous attempt's test results more often failed other tests that the reference fix passes, which shrinks their apparent gains. Fifth, outcomes varied far more between failures than between conditions, and we give the sample sizes a confirmatory study would need.

cs.SE↗

arXiv2026

Component Translation and Judge-Labelled Refusal: A Multilingual Safety Evaluation

Whole-prompt translation evaluates a request in different languages but does not compare which component carries a language. We introduce a paired reversal: translate the complete request component or the complete operation component into the same target language, holding behaviour, English frame and translated-span count fixed. Component identity and non-English text quantity change together. Assigning English, Chinese and Spanish independently to the two components yields nine configurations on 190 of 300 sampled harmful behaviours retained after consensus annotation and automatic gating. Refusal scored by language-model judges is the primary outcome; unsafe compliance is reported separately. On Gemini 2.5 Flash and DeepSeek, three refusal swaps are about 5.3 percentage points and the fourth is -0.5; none survives the pre-specified six-test Holm correction. In an exploratory extension, Mistral Small 4 has a 13.7-point Spanish swap (95% interval [8.4, 18.9]), which survives the twelve-test Holm sensitivity correction under the primary judge, and 12.1 points under a second judge. GPT-5.4-mini estimates are within 2 points of zero. Component assignment supplies an additional evaluation dimension. Evaluations should record component languages and translated text quantity, comparing matched behaviours at equal translated-span count while interpreting refusal and unsafe compliance separately.

cs.CL↗

arXiv2026

Style Over Substance: Content-Invariant Wrappers Flip LLM Safety-Judge Verdicts

Automatic safety judges -- for example Llama Guard, or GPT-4o given a grading prompt -- produce the numbers behind reported jailbreak success rates, defense evaluations, and safety leaderboards. We ask whether these judges grade what a reply contains or how it sounds. We place a fixed piece of text before or after a model reply (a style wrapper) and leave the reply itself unchanged byte-for-byte, so any change in verdict is an error of the judge. Over 600 JailbreakBench replies (300 harmful, 300 refusals), 9 wrappers, and 8 judges, using paired tests and a measured per-judge repeat-disagreement rate (the noise floor), we find that most judges barely move but specific judges have specific, cheap blind spots. A token refusal ("I can't help with that") placed before the unchanged harmful text flips 19.9% of GPT-4o-mini's correct "unsafe" verdicts to "safe" (noise floor 0.5%; 18.2% under majority-of-three re-scoring), yet moves Claude only 0.4%. The deployed Llama Guard 4 guard model is evaded by an "educational course" preface on 12.3% of harmful replies, and the flips persist across repeats, while a second deployed guard (gpt-oss-safeguard-20b) is immune. Changing only the grading prompt to ignore tone cuts the GPT-4o-mini attack roughly tenfold on the same model, so the weakness lives in the judge, not in the content. Pure tone-only decoration that adds no new claim still flips up to 2.1% of verdicts on the vulnerable judges, several times their noise floors. On this artifact set the judges also disagree about which target model is safest, and a bootstrap shows the ranking is unstable before any wrapper is added. We release the data, wrappers, code, and per-verdict labels.

cs.AI↗

arXiv2026

A Very Big Video Reasoning Suite

Rapid progress in video models has largely focused on visual quality, leaving their reasoning capabilities underexplored. Video reasoning grounds intelligence in spatiotemporally consistent visual environments that go beyond what text can naturally capture, enabling intuitive reasoning over spatiotemporal structure such as continuity, interaction, and causality. However, systematically studying video reasoning and its scaling behavior is hindered by the lack of large-scale training data. To address this gap, we introduce the Very Big Video Reasoning (VBVR) Dataset, an unprecedentedly large-scale resource spanning 200 curated reasoning tasks following a principled taxonomy and over one million video clips, approximately three orders of magnitude larger than existing datasets. We further present VBVR-Bench, a verifiable evaluation framework that moves beyond model-based judging by incorporating rule-based, human-aligned scorers, enabling reproducible and interpretable diagnosis of video reasoning capabilities. Leveraging the VBVR suite, we conduct one of the first large-scale scaling studies of video reasoning and observe early signs of emergent generalization to unseen reasoning tasks. Together, VBVR lays a foundation for the next stage of research in generalizable video reasoning. The data, benchmark toolkit, and models are publicly available at https://video-reason.com/?v=vbvr .

cs.CV↗

arXiv2026

Probing Perceptual Constancy in Large Vision-Language Models

Perceptual constancy is the ability to maintain stable perceptions of objects despite changes in sensory input, such as variations in distance, angle, or lighting. This ability is crucial for visual understanding in a dynamic world. Here, we explored such ability in current Vision Language Models (VLMs). In this study, we evaluated 155 VLMs using 236 experiments across three domains: color, size, and shape constancy. The experiments included single-image and video adaptations of classic cognitive tasks, along with novel tasks in in-the-wild conditions. We found significant variability in VLM performance across these domains, with model performance in shape constancy clearly dissociated from that of color and size constancy.

cs.CV↗