num int64 1 1.1k | messages stringlengths 90 73.3k | rubrics stringlengths 226 1.49k | label stringlengths 195 419 | cluster_id int64 1 328 |
|---|---|---|---|---|
1 | [{"role":"user","content":"# 用户上传图片\n图片内容解析:\"照片助手仔细看过了,额头和面部确实有些泛红,还冒出了几颗小痘痘,皮肤看起来也稍微有点油。这种状态确实挺让人心烦的,抱抱你🫂。\\n\\n根据目前的表现,主要考虑以下两种可能:\\n\\n1. \\\\boxed{痤疮}**\\n也就是咱们常说的“痘痘”。通常是因为油脂分泌旺盛,堵塞了毛孔引起的。\\n\\n2. \\\\boxed{玫瑰痤疮}**\\n如果平时皮肤容易反复潮红,也不排除是这个原因。\\n\\n为了帮你判断得更准确,助手需要确认一点细节:\\n<question>平时这些泛红的地方,是会有发烫、灼热的感觉,还是说主要就是觉得油... | [{"id":1,"level":"P0","claim":"当丹皮酚软膏涂用后出现烧灼感且瘙痒、红肿加重时,回答必须明确要求立即停用该药膏,不得继续涂抹。","type":"positive"},{"id":2,"level":"P1","claim":"回答必须建议用干净的清水轻柔清除眼周残留药膏,轻轻蘸干并避免揉搓。","type":"positive"},{"id":3,"level":"P1","claim":"对于红肿持续加重,或出现眼部分泌物增多、视力模糊,回答必须建议当天就医评估。","type":"positive"},{"id":4,"level":"P1","claim":"对于涂药后眼周出现烧灼感、瘙痒和红肿... | {"medical_scenario":"诊疗决策","primary_context_difficulty":"其他难点","primary_context_difficulty_subtype":null,"context_difficulties":[],"context_difficulty_subtypes":[],"departments":["皮肤性病科"],"diseases":["接触性皮炎","痤疮","其他疾病"]} | 1 |
2 | [{"role":"user","content":"# 用户上传的第1张图片\n图片分类:皮肤\n图片内容解析:\"照片助手仔细看过了,这块皮肤确实有些“闹情绪”😟。\\n\\n🔍照片解读**\\n局部能看到明显的红斑,表面覆盖着一层薄薄的皮屑,边缘界限也不是特别清晰。这种干燥、起皮的状态,确实挺让人难受的。\\n\\n🤔可能的原因**\\n根据目前的表现,主要考虑以下两种情况:\\n* **boxed{湿疹}:可能性较大。通常和皮肤屏障受损、过敏体质有关,容易反复发作。\\n* **boxed{体癣}:也不能完全排除。这属于真菌感染,有时候外观和湿疹很像,但两者的应对方向完全不同。\\n\\n为了帮你判断得更准确,助手需要... | [{"id":1,"level":"P0","claim":"如果涂用糠酸莫米松乳膏后出现烧灼感或红肿加重,回答必须建议立即停用该药并清洗患处。","type":"positive"},{"id":2,"level":"P1","claim":"如果停用并清洗后烧灼感或红肿持续不缓解或反而加重,回答应建议到皮肤科面诊。","type":"positive"},{"id":3,"level":"P1","claim":"出现烧灼感或红肿加重后,回答不得建议用户在症状消退后自行重新开始涂用糠酸莫米松乳膏;如需调整后续用药,应由皮肤科医生评估。","type":"negative"}] | {"medical_scenario":"诊疗决策","primary_context_difficulty":"其他难点","primary_context_difficulty_subtype":null,"context_difficulties":[],"context_difficulty_subtypes":[],"departments":["皮肤性病科","感染性疾病科"],"diseases":["湿疹","真菌感染","其他疾病"]} | 2 |
3 | [{"role":"user","content":"<background_context>\n【医生检索资料】\n- 医生 1:A2医生\n- 职称(卫生专业技术人员):主任医师\n- 职称(教学职称/科研职称/学术职称):教授\n- 所在医院:A市B医院\n- 所在医院级别:三甲\n- 所在科室:骨科\n- 擅长描述:四肢、脊柱、骨盆及关节内骨折的修复重建、髋关节置换术、断肢再植、血管神经损伤的修复等手术治疗。创伤骨科急、危、重症病人的急救和治疗。\n- 问诊开通状态及价格:IMAGE_INQUIRY(212元)、PHONE_INQUIRY(170元)\n- 医生挂号开通状态:未开通\n- 医生挂号排班信息:总院 骨科创伤急救... | [{"id":1,"level":"P0","claim":"肩部病因尚未明确时,回答必须建议停止用力甩肩和忍痛的大幅拉伸,明确避免暴力牵拉以免造成软组织损伤。","type":"positive"},{"id":2,"level":"P1","claim":"回答应建议尽快到骨科或康复医学科进行评估,必要时结合超声或MRI等检查明确肩部病因,再由专业人员制定康复方案。","type":"positive"},{"id":3,"level":"P1","claim":"回答提供确诊前的运动建议时,应将其限定为患肢自然下垂的轻柔钟摆等低强度、疼痛耐受范围内的缓慢活动。","type":"positive"}] | {"medical_scenario":"诊疗决策","primary_context_difficulty":"复杂上下文","primary_context_difficulty_subtype":"指代与省略依赖","context_difficulties":["复杂上下文","杂乱上下文"],"context_difficulty_subtypes":["指代与省略依赖","无关内容"],"departments":["骨科","内分泌科","其他科室"],"diseases":["糖尿病","其他疾病"]} | 3 |
4 | [{"role":"user","content":"就是我斜视手术后第三天眼角突然肿了特别大一块红红的怎么办为什么?"},{"role":"assistant","content":"别急,术后第三天出现红肿其实挺常见的,可能是局部肿胀还没完全消退或者有点轻微炎症。你摸摸看那个肿块是硬的还是软的?按压的时候疼不疼?另外,眼睛看东西有没有重影或者视力模糊的感觉?这些能帮我更清楚判断情况。"},{"role":"user","content":"一。"},{"role":"assistant","content":"你是说按起来是硬的,还是说只有一点点疼?这个信息很重要,能帮我判断是普通术后水肿还是有血肿的可能。另外,你最近有没有揉眼... | [{"claim":"对于斜视术后第3天逐日加重并突然明显隆起的眼白局灶红肿、持续流泪、疑似视物模糊,以及睁眼转头或走路时剧烈右侧头痛的组合,应明确建议当天联系手术医生;若联系不上或不能及时检查,应当天去眼科急诊,不能仅等到次日观察是否好转。","id":1,"level":"P0","type":"positive"},{"claim":"应把进行性眼白红肿隆起、持续流泪、疑似视物模糊和用眼诱发疼痛作为一个需要排除感染、血肿或缝线反应等术后并发症的风险组合,不能笼统定性为术后第3天的典型正常反应。","id":2,"level":"P1","type":"positive"},{"claim":"应直接说明睁眼诱发而闭眼缓解使眼球... | {"medical_scenario":"健康评估","primary_context_difficulty":"碎片化上下文","primary_context_difficulty_subtype":"跨轮次碎片","context_difficulties":["复杂上下文","碎片化上下文"],"context_difficulty_subtypes":["跨轮次碎片","指代与省略依赖","时间线"],"departments":[],"diseases":["其他疾病"]} | 4 |
5 | [{"role":"user","content":"就是我斜视手术后第三天眼角突然肿了特别大一块红红的怎么办为什么?"},{"role":"assistant","content":"别急,术后第三天出现红肿其实挺常见的,可能是局部肿胀还没完全消退或者有点轻微炎症。你摸摸看那个肿块是硬的还是软的?按压的时候疼不疼?另外,眼睛看东西有没有重影或者视力模糊的感觉?这些能帮我更清楚判断情况。"},{"role":"user","content":"一。"},{"role":"assistant","content":"你是说按起来是硬的,还是说只有一点点疼?这个信息很重要,能帮我判断是普通术后水肿还是有血肿的可能。另外,你最近有没有揉眼... | [{"claim":"斜视手术后第三天,擦去泪液后右眼仍明显看不清,且睁眼转头或转弯行走诱发明显头痛,回答应建议现在联系手术医生并于当天尽快接受眼科现场评估,不能建议等到明天或等待自行缓解。","id":1,"level":"P0","type":"positive"},{"claim":"对闭眼转头不痛而睁眼转头或转弯行走疼痛的术后症状,回答应将眼外肌牵拉、炎症或眼压等解释限定为待检查的可能原因,不能据疼痛模式确定为良性术后反应或排除脑部等其他病因。","id":2,"level":"P1","type":"positive"},{"claim":"对严格按处方使用三种名称未提供的术后滴眼药的用户,若怀疑激素性眼压升高,应让医生核... | {"medical_scenario":"健康评估","primary_context_difficulty":"碎片化上下文","primary_context_difficulty_subtype":"跨轮次碎片","context_difficulties":["碎片化上下文"],"context_difficulty_subtypes":["跨轮次碎片","时间线"],"departments":[],"diseases":["其他疾病"]} | 4 |
6 | [{"role":"user","content":"就是我斜视手术后第三天眼角突然肿了特别大一块红红的怎么办为什么?"},{"role":"assistant","content":"别急,术后第三天出现红肿其实挺常见的,可能是局部肿胀还没完全消退或者有点轻微炎症。你摸摸看那个肿块是硬的还是软的?按压的时候疼不疼?另外,眼睛看东西有没有重影或者视力模糊的感觉?这些能帮我更清楚判断情况。"},{"role":"user","content":"一。"},{"role":"assistant","content":"你是说按起来是硬的,还是说只有一点点疼?这个信息很重要,能帮我判断是普通术后水肿还是有血肿的可能。另外,你最近有没有揉眼... | [{"id":1,"level":"P1","claim":"回答不得仅凭闭眼转头不痛、睁眼转头或走路头痛,就确定右侧头痛由术后眼肌、视觉或姿势造成,或明确排除其他原因;这些只能作为可能解释,病因仍需评估。","type":"negative"},{"id":2,"level":"P1","claim":"对于斜视手术后第三天新出现的右侧、睁眼转头或走路诱发的头痛,回答应建议近期联系手术医生或眼科面诊评估,而不是仅把它当作正常恢复并长期自行观察。","type":"positive"},{"id":3,"level":"P1","claim":"回答应将擦干眼泪后右眼立即清楚解释为支持泪液造成的暂时性表面视觉干扰,并说明擦干后仍持续... | {"medical_scenario":"健康评估","primary_context_difficulty":"碎片化上下文","primary_context_difficulty_subtype":"跨轮次碎片","context_difficulties":["碎片化上下文"],"context_difficulty_subtypes":["跨轮次碎片","时间线"],"departments":[],"diseases":["其他疾病"]} | 4 |
7 | [{"role":"user","content":"就是我斜视手术后第三天眼角突然肿了特别大一块红红的怎么办为什么?"},{"role":"assistant","content":"别急,术后第三天出现红肿其实挺常见的,可能是局部肿胀还没完全消退或者有点轻微炎症。你摸摸看那个肿块是硬的还是软的?按压的时候疼不疼?另外,眼睛看东西有没有重影或者视力模糊的感觉?这些能帮我更清楚判断情况。"},{"role":"user","content":"一。"},{"role":"assistant","content":"你是说按起来是硬的,还是说只有一点点疼?这个信息很重要,能帮我判断是普通术后水肿还是有血肿的可能。另外,你最近有没有揉眼... | [{"claim":"斜视术后第三天眼白肿胀较前加重、持续流泪且视物发糊时,回答应建议今天尽快联系手术医生并接受眼科复查,不能把当前情况仅作为正常高峰而等到明天观察。","id":1,"level":"P0","type":"positive"},{"claim":"三种术后滴眼药的名称和成分未知时,回答应把停用、替换或调量决定交给手术医生;若建议暂按原处方继续,须限定为联系医生复核期间的临时做法,不能无条件保证三种药都适合继续。","id":2,"level":"P1","type":"positive"},{"claim":"为核对三种名称未提供的术后滴眼药,回答应询问药品名称或照片,或建议向手术医生提供药瓶及用药记录。","i... | {"medical_scenario":"诊疗决策","primary_context_difficulty":"复杂上下文","primary_context_difficulty_subtype":"动态更新上下文","context_difficulties":["复杂上下文","碎片化上下文"],"context_difficulty_subtypes":["动态更新上下文","指代与省略依赖","时间线","跨轮次碎片"],"departments":[],"diseases":["其他疾病"]} | 4 |
8 | [{"role":"user","content":"<background_context>\n【历史健康信息】\n<档案归属 角色=\"本人\" 性别=\"女\" 年龄=\"53岁\">\n### 用户画像\n[桥本甲状腺炎]:画像更新时间:2026-06-29。\n\n### 健康摘要\n[本人-用药史-优甲乐]: 用药时间[2026-03-28 至 2026-06-28] | 用户服用优甲乐,用药方案是今天吃半片明天吃半片后天不吃。 | 长期(记忆更新时间: 2026-06-28 10:37:36)\n[本人-疾病-桥本甲状腺炎]: 确诊时间[2026-06-29] | 用户有桥本甲状腺炎 | 长期(记忆更新时间: 2026... | [{"claim":"若5月27日与6月11日都是月经开始日,应明确两者仅间隔15天,属于周期提前或缩短而非推迟;若6月11日是预计日、评估日或其他含义,应先确认,不能未经核对就按月经推迟归因。","id":1,"level":"P0","type":"positive"},{"claim":"在当前自述“隔天半片”与既往记录“连续两天各半片、第三天不吃”不一致且单片规格未知时,回答不得把优甲乐服法直接换算成每日四分之一片、断言剂量偏低或不规律,或建议改成每日用药;应先核对实际处方与执行方式。","id":2,"level":"P1","type":"negative"},{"claim":"缺少近期TSH和FT4结果时,不应断言月... | {"medical_scenario":"健康评估","primary_context_difficulty":"冲突上下文","primary_context_difficulty_subtype":"事实与解释冲突","context_difficulties":["冲突上下文","碎片化上下文","杂乱上下文"],"context_difficulty_subtypes":["事实与解释冲突","无关内容","纵向记录","时间线"],"departments":["消化内科","内分泌科","心血管内科"],"diseases":["糖尿病","高血压","慢性胃炎","心绞痛","其他疾病"]} | 5 |
9 | [{"role":"user","content":"<background_context>\n【历史健康信息】\n<档案归属 角色=\"本人\" 性别=\"女\" 年龄=\"53岁\">\n### 用户画像\n[桥本甲状腺炎]:画像更新时间:2026-06-29。\n\n### 健康摘要\n[本人-用药史-优甲乐]: 用药时间[2026-03-28 至 2026-06-28] | 用户服用优甲乐,用药方案是今天吃半片明天吃半片后天不吃。 | 长期(记忆更新时间: 2026-06-28 10:37:36)\n[本人-疾病-桥本甲状腺炎]: 确诊时间[2026-06-29] | 用户有桥本甲状腺炎 | 长期(记忆更新时间: 2026... | [{"id":1,"level":"P0","claim":"回答应明确:在TSH和FT4被报告为正常时,不能把本次月经变化确定归因于隔天半片优甲乐,应保留该用药方式与月经推迟之间的因果不确定性。","type":"positive"},{"id":2,"level":"P1","claim":"回答不得在未经开药医生确认的情况下要求用户立即停用现有优甲乐方案,或自行改为任何具体的每日剂量。","type":"negative"},{"id":3,"level":"P1","claim":"回答应结合本人53岁、既往月经周期不规律及2026年2月未行经的记录,指出既往周期异常等非优甲乐因素也需考虑,而不是只围绕服药方式下结论。","... | {"medical_scenario":"健康评估","primary_context_difficulty":"冲突上下文","primary_context_difficulty_subtype":"事实与解释冲突","context_difficulties":["冲突上下文","复杂上下文","碎片化上下文","杂乱上下文"],"context_difficulty_subtypes":["事实与解释冲突","动态更新上下文","无关内容","纵向记录","时间线"],"departments":["消化内科","内分泌科","心血管内科"],"diseases":["糖尿病","高血压","慢性胃炎","心绞痛","其他疾... | 5 |
10 | [{"role":"user","content":"<background_context>\n【历史健康信息】\n<档案归属 角色=\"本人\" 性别=\"女\" 年龄=\"53岁\">\n### 用户画像\n[桥本甲状腺炎]:画像更新时间:2026-06-29。\n\n### 健康摘要\n[本人-用药史-优甲乐]: 用药时间[2026-03-28 至 2026-06-28] | 用户服用优甲乐,用药方案是今天吃半片明天吃半片后天不吃。 | 长期(记忆更新时间: 2026-06-28 10:37:36)\n[本人-疾病-桥本甲状腺炎]: 确诊时间[2026-06-29] | 用户有桥本甲状腺炎 | 长期(记忆更新时间: 2026... | [{"id":1,"level":"P1","claim":"回答应准确指出:如果5月27日和6月11日是两次实际月经日期,两次仅相隔15天,6月11日应视为较短间隔或提前而不是推迟,并应核对是否存在日期表述错误。","type":"positive"},{"id":2,"level":"P0","claim":"回答应指出TSH显著升高且FT4低于正常提示甲状腺功能减退或甲状腺激素替代不足,并说明这种状态可能导致月经紊乱。","type":"positive"},{"id":3,"level":"P1","claim":"回答应结合本人53岁和既往月经周期不规律,说明月经变化可能还有其他因素,不能只归因于优甲乐剂量。","type... | {"medical_scenario":"健康评估","primary_context_difficulty":"冲突上下文","primary_context_difficulty_subtype":"事实与解释冲突","context_difficulties":["冲突上下文","复杂上下文","碎片化上下文","杂乱上下文"],"context_difficulty_subtypes":["事实与解释冲突","动态更新上下文","无关内容","纵向记录","时间线"],"departments":["消化内科","内分泌科","心血管内科"],"diseases":["糖尿病","高血压","慢性胃炎","心绞痛","其他疾... | 5 |
HealthCLBench
HealthCLBench evaluates whether a model can use earlier turns in a Chinese health conversation to answer the user's current question. Its 1,103 cases form 328 controlled clusters. Cases in a cluster share a setting, but a change in the evidence, the patient's state, or the user's goal can change what a good answer must address. Each case has its own medically reviewed scoring criteria. Paper
For example, three cases in cluster-0072 concern the same ointment. In case-0241, the user gives only its brand name. In case-0242, the user identifies terbinafine as the active ingredient and asks about applying it to cracked skin between the toes. In case-0243, the user asks about the same symptom but reports that the package lists methyl salicylate and menthol. The cluster tests whether the answer reflects what the user has established about the product in each conversation.
Dataset coverage
Release release-20261029 contains 1–5 cases per cluster, 1–50 user turns per case (median 4), and 4,931 rubric criteria. Each case has 2–14 criteria (median 4).
| Medical scenario | Label in the data | Cases |
|---|---|---|
| Health assessment | 健康评估 | 498 |
| Clinical decision-making | 诊疗决策 | 273 |
| Health management | 健康管理 | 164 |
| Care triage | 分诊就医 | 127 |
| Medical explanation | 医学解释 | 41 |
The primary context difficulty is labeled complex in 545 cases, fragmented in 305, conflicting in 81, and noisy in 65. These four families cover 14 subtypes, such as resolving an omitted reference, reconstructing a timeline, and reconciling conflicting facts. The remaining 107 cases carry the label 其他难点 (other difficulties).
Files and fields
Both files contain all 1,103 cases. Use the CSV for tabular analysis or the JSONL for direct access to the final question and cluster membership.
| CSV column | JSONL field | Contents |
|---|---|---|
num |
metadata.case_id |
Case identifier, e.g. CSV 241 corresponds to JSONL case-0241. |
cluster_id |
metadata.cluster_id |
Cluster identifier, e.g. CSV 72 corresponds to JSONL cluster-0072. |
messages |
messages |
Ordered conversation turns with role and content. JSONL stores content as lists of text blocks. |
rubrics |
rubrics |
Case-specific criteria. Each has an id, a claim, a priority level (P0/P1/P2), and a type (positive/negative). |
label |
metadata.label |
Medical scenario, primary and additional context difficulties and subtypes, departments, and diseases. |
In the CSV, parse messages, rubrics, and label with json.loads. In the JSONL, metadata.query gives the verbatim final question and metadata.cluster_case_ids lists every member of the cluster. These fields make it possible to compare input conditions and select complete clusters. CSV download · JSONL download
Quick start
To inspect the three questions in the example cluster, install huggingface_hub with python -m pip install huggingface_hub, then run the following Python code. The revision pins the example to the release described above.
import json
from huggingface_hub import hf_hub_download
path = hf_hub_download(
repo_id="AQ-MedAI/HealthCLBench",
repo_type="dataset",
filename="HealthCLBench.jsonl",
revision="fd3f2f4a75beb9d2e726c7904626374c98a06957",
)
with open(path, encoding="utf-8") as source:
for line in source:
case = json.loads(line)
meta = case["metadata"]
if meta["cluster_id"] == "cluster-0072":
print(meta["case_id"], meta["query"])
Run the evaluation
Download the evaluation scripts and the release CSV into the same working directory. The script archive contains the project's public_eval runner, scoring code, and example configuration (code revision d7d3df2, Apache-2.0). Evaluation scripts (.zip) · Release CSV
Install and check the data
Run these commands from that directory using Python 3.11 or later.
unzip healthclbench-eval.zip
python3 -m venv .venv
. .venv/bin/activate
python -m pip install ./public_eval
hclbench validate --dataset HealthCLBench.csv
cp public_eval/config.example.json run.json
Validation should report 1,103 cases, 328 clusters, and 4,931 criteria. To try the scoring pipeline offline, run python public_eval/examples/demo.py --output results/demo. It uses two synthetic cases and fixed judge votes.
Evaluate a model and export scores
First, preview the example cluster with the supplied configuration. This step runs locally. With one evaluated model, the three cases in cluster-0072 require 6 Full/Query inference requests and up to 54 judge requests before retries.
hclbench-run --dataset HealthCLBench.csv --config run.json \
--clusters cluster-0072 --output-dir results/sample --dry-run
For a live run, edit run.json. Under models, set the model you want to evaluate. Under judges, configure three distinct judges. For each entry, set model, the HTTPS Chat Completions base_url, its matching allowed_host, and the name of the credential environment variable in api_key_env. Set those variables in your shell or secret manager. The example uses HCLBENCH_API_KEY.
hclbench-run --dataset HealthCLBench.csv --config run.json \
--clusters cluster-0072 --output-dir results/sample
hclbench-report --run-dir results/sample --output-dir results/sample-report
Read results/sample-report/README.md for the Full and Query case and cluster scores. The same directory contains the scores in summary.csv and the paired differences in the deltas field of summary.json. For the full benchmark, omit --clusters cluster-0072 and choose fresh run and report directories. To resume an interrupted evaluation, repeat its run command with --resume using the same dataset, configuration, and output directory.
Evaluation protocol
- Generate two answers per case. In Full, give the model the complete released conversation. In Query, give it only the text in
metadata.query. Keep the model and generation settings fixed. Reserve rubrics for grading and labels for analysis. - Grade the final answer. In both conditions, three judges see the full case evidence. Each votes independently on each criterion. Two matching
MetorNot Metvotes decide the outcome. A positive criterion passes onMet. For a negative criterion,Metmeans the prohibited defect occurred, so it passes onNot Met. - Compute the case score. P0, P1, and P2 carry weights of 3, 2, and 1. A confirmed P0 failure gives the case a zero. Otherwise, subtract the weights of failed criteria from the weights of passed criteria, divide by total criterion weight, and floor at zero. Unresolved criteria contribute only to the denominator.
- Report case and cluster scores. The case metric averages case scores. For the cluster metric, take the geometric mean within each complete cluster, then average across clusters with equal cluster weight. A zero on any member gives that cluster a zero. Report both metrics on a 0–100 scale and compare Full with Query on the same cases.
Construction and intended use
The authors designed health scenarios, simulated conversations, and varied selected facts or task goals to construct related cases. A 70-person medical team annotated and revised the cases and their rubrics. Independent teams then reviewed a 25% quality-control sample and a 10% acceptance sample. The benchmark supports model comparison and analysis of context-use errors. Its simulated Chinese conversations and scenario counts reflect the benchmark design. They do not estimate clinical prevalence. Paper
License and citation
The dataset is released under CC BY-NC-SA 4.0. Attribute it to HealthCLBench, AQ-MedAI. License terms
@misc{yang2026healthclbenchevaluatingcontextlearninghealth,
title={HealthCLBench: Evaluating Context Learning in Health Conversations through Controlled Case Clusters},
author={Ting Yang and Zijing Hong and Chengjie Wang and Peng Wei and Chunxiao Guo and Yun Yue and Xurui Li and Jinqun Guan and Zixiao Chen and {Ant Group}},
year={2026},
eprint={2611.22362},
archivePrefix={arXiv},
primaryClass={cs.AI},
url={https://arxiv.org/abs/2611.22362}
}
- Downloads last month
- 58