{"id":"arxiv:2502.15840","title":"Vending-Bench: A Benchmark for Long-Term Coherence of Autonomous Agents","date":"2025-02-21","section":"Agents, open-endedness, AGI","quality":-0.282468,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/3e821cc90ae1748114d12211bc9209146fea5b544c192e41be419dd6dc9ee2b4/","source_url":"https://arxiv.org/abs/2502.15840","text_available":true,"authors":["Axel Backlund","Lukas Petersson"],"outgoing":[],"incoming":[{"source":"arxiv:2606.18543","target":"arxiv:2502.15840","evidence":"id","id":"arxiv:2606.18543","title":"CEO-Bench: Can Agents Play the Long Game?","url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/696d7ac1dfd7539fbe8147ed713574b8eb5a7400a35e96805c7f0c9add99165c/"}],"text_parts":[{"start":0,"end":15988,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/3e821cc90ae1748114d12211bc9209146fea5b544c192e41be419dd6dc9ee2b4/part-1.html"},{"start":15988,"end":31800,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/3e821cc90ae1748114d12211bc9209146fea5b544c192e41be419dd6dc9ee2b4/part-2.html"}],"contents":[{"title":"Vending-Bench: A Benchmark for Long-Term Coherence of Autonomous Agents","offset":0,"level":1,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/3e821cc90ae1748114d12211bc9209146fea5b544c192e41be419dd6dc9ee2b4/part-1.html#text"},{"title":"Abstract","offset":75,"level":6,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/3e821cc90ae1748114d12211bc9209146fea5b544c192e41be419dd6dc9ee2b4/part-1.html#text"},{"title":"1 Introduction","offset":1525,"level":2,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/3e821cc90ae1748114d12211bc9209146fea5b544c192e41be419dd6dc9ee2b4/part-1.html#text"},{"title":"2 Method","offset":4292,"level":2,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/3e821cc90ae1748114d12211bc9209146fea5b544c192e41be419dd6dc9ee2b4/part-1.html#text"},{"title":"2.1 Agent implementation","offset":4305,"level":3,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/3e821cc90ae1748114d12211bc9209146fea5b544c192e41be419dd6dc9ee2b4/part-1.html#text"},{"title":"2.2 Task environment","offset":5638,"level":3,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/3e821cc90ae1748114d12211bc9209146fea5b544c192e41be419dd6dc9ee2b4/part-1.html#text"},{"title":"2.2.1 Simulating supplier communication","offset":7512,"level":4,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/3e821cc90ae1748114d12211bc9209146fea5b544c192e41be419dd6dc9ee2b4/part-1.html#text"},{"title":"2.2.2 Simulating customer purchases","offset":8727,"level":4,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/3e821cc90ae1748114d12211bc9209146fea5b544c192e41be419dd6dc9ee2b4/part-1.html#text"},{"title":"2.3 Environment configuration","offset":9661,"level":3,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/3e821cc90ae1748114d12211bc9209146fea5b544c192e41be419dd6dc9ee2b4/part-1.html#text"},{"title":"2.4 Scoring","offset":10552,"level":3,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/3e821cc90ae1748114d12211bc9209146fea5b544c192e41be419dd6dc9ee2b4/part-1.html#text"},{"title":"2.5 Human baseline","offset":10949,"level":3,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/3e821cc90ae1748114d12211bc9209146fea5b544c192e41be419dd6dc9ee2b4/part-1.html#text"},{"title":"3 Results and discussion","offset":11375,"level":2,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/3e821cc90ae1748114d12211bc9209146fea5b544c192e41be419dd6dc9ee2b4/part-1.html#text"},{"title":"3.1 Overview","offset":11404,"level":3,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/3e821cc90ae1748114d12211bc9209146fea5b544c192e41be419dd6dc9ee2b4/part-1.html#text"},{"title":"3.2 Primary models","offset":12630,"level":3,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/3e821cc90ae1748114d12211bc9209146fea5b544c192e41be419dd6dc9ee2b4/part-1.html#text"},{"title":"3.2.1 Comparison of top models","offset":14834,"level":4,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/3e821cc90ae1748114d12211bc9209146fea5b544c192e41be419dd6dc9ee2b4/part-1.html#text"},{"title":"3.2.2 Trace analysis","offset":17989,"level":4,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/3e821cc90ae1748114d12211bc9209146fea5b544c192e41be419dd6dc9ee2b4/part-2.html#text"},{"title":"3.3 Secondary models","offset":21467,"level":3,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/3e821cc90ae1748114d12211bc9209146fea5b544c192e41be419dd6dc9ee2b4/part-2.html#text"},{"title":"3.3.1 Trace analysis","offset":22889,"level":4,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/3e821cc90ae1748114d12211bc9209146fea5b544c192e41be419dd6dc9ee2b4/part-2.html#text"},{"title":"3.4 Comparison to human baseline","offset":24451,"level":3,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/3e821cc90ae1748114d12211bc9209146fea5b544c192e41be419dd6dc9ee2b4/part-2.html#text"},{"title":"3.5 Experiment variations","offset":25512,"level":3,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/3e821cc90ae1748114d12211bc9209146fea5b544c192e41be419dd6dc9ee2b4/part-2.html#text"},{"title":"3.5.1 Environment configuration","offset":25543,"level":4,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/3e821cc90ae1748114d12211bc9209146fea5b544c192e41be419dd6dc9ee2b4/part-2.html#text"},{"title":"3.5.2 Agent memory","offset":26597,"level":4,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/3e821cc90ae1748114d12211bc9209146fea5b544c192e41be419dd6dc9ee2b4/part-2.html#text"},{"title":"3.6 Is the problem long input length?","offset":27794,"level":3,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/3e821cc90ae1748114d12211bc9209146fea5b544c192e41be419dd6dc9ee2b4/part-2.html#text"},{"title":"4 Conclusion","offset":29338,"level":2,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/3e821cc90ae1748114d12211bc9209146fea5b544c192e41be419dd6dc9ee2b4/part-2.html#text"},{"title":"Appendix A Appendix","offset":30812,"level":2,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/3e821cc90ae1748114d12211bc9209146fea5b544c192e41be419dd6dc9ee2b4/part-2.html#text"},{"title":"Instructions for reporting errors","offset":30836,"level":2,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/3e821cc90ae1748114d12211bc9209146fea5b544c192e41be419dd6dc9ee2b4/part-2.html#text"}],"text_code_points":31800,"source_markdown":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/3e821cc90ae1748114d12211bc9209146fea5b544c192e41be419dd6dc9ee2b4/source.md"}