Reflexion: Language Agents with Verbal Reinforcement Learning · HackerPapers discussions

Reflexion: Language Agents with Verbal Reinforcement Learning — HN submission

Mon, 02 Mar 2026 17:44:23 GMT

Show HN: Benchmarking the Keep memory system with LoCoMo

Summary: keep (https://github.com/hughpyle/keep/, MIT-licensed) is a skills practice wrapped around an implementation of "memory for AI agents". The practice is this: repeated reflection on means and outcomes, so that skillful action improves over time. But the raw implementation of memory is its foundation. Without working memory, you can't iterate. Similarly, without benchmarks, you can't tell what works. Today we're publishing results for the LoCoMo benchmark. Scores: 76.2% overall (weighted average) Single-hop 86.2% (841 questions) Temporal 68.5% (321 questions) Multi-hop 64.2% (282…

1 point; 0 discussion replies on HN. Paper and source history. A submission is not an endorsement.

Reflexion: Language Agents with Verbal Reinforcement Learning — mentioned by ipsum2

Thu, 03 Oct 2024 23:32:26 GMT

No. CoT has been around for several years (Jan 2022) https://arxiv.org/abs/2201.11903. And so has Reflection (March 2023) https://arxiv.org/abs/2303.11366. This approach that was taken by Reflection is nothing new.

Paper and discussion history. A mention is not an endorsement.

Reflexion: Language Agents with Verbal Reinforcement Learning — mentioned by unoti

Mon, 25 Sep 2023 17:40:48 GMT

I think you're 100% on the right track here. The key is memory, loops, and maybe a few other things like external interfaces which are just plain code and not deep learning voodoo. Many things do indeed run LLM's in a loop and attach external sources. See for example AutoGPT, the ReAct paper[1], and the Reflexion paper[2]. ReAct one line summary: This is about giving the machine tools that are external interfaces, integrating those with the llm and teaching it how to use those tools with a few examples, and then letting it run the show to fulfill the user's ask/question and using the tools…

Paper and discussion history. A mention is not an endorsement.

Reflexion: Language Agents with Verbal Reinforcement Learning — mentioned by avindroth

Fri, 25 Aug 2023 03:28:05 GMT

Reflexion can serve as good reference here: https://arxiv.org/abs/2303.11366 Essentially, AI’s output is fed into a checker, whose output is fed back into the AI for “reflexion”. Then the AI often corrects (leading to noticeable improvement in GPT-4 perf).

Paper and discussion history. A mention is not an endorsement.

Reflexion: Language Agents with Verbal Reinforcement Learning — mentioned by famouswaffles

Tue, 11 Jul 2023 19:47:09 GMT

I don't know how we'd go about comparing reflection to human baselines but GPT-4 can reflect just fine https://arxiv.org/abs/2303.11366 GPT-4's emotional intelligence seems to be very high by the looks of things https://arxiv.org/abs/2304.11490 Empathy =/ Intelligence

Paper and discussion history. A mention is not an endorsement.

Reflexion: Language Agents with Verbal Reinforcement Learning — mentioned by lhl

Tue, 11 Jul 2023 14:12:14 GMT

Since I've been on a AI code-helper kick recently. According to the post, Claude 2 now 71.2%, a significant upgrade from 1.3 (56.0%). (Found in model card: pass@1) For comparison: * GPT-4 claims 85.4 on HumanEval, in a recent paper https://arxiv.org/pdf/2303.11366.pdf GPT-4 was tested at 80.1 pass@1 and 91 pass@1 using their Reflexion technique. They also include MBPP and Leetcode Hard benchmark comparisons * WizardCoder, a StarCoder fine-tune is one of the top open models, scoring a 57.3 pass@1, model card here: https://huggingface.co/WizardLM/WizardCoder-15B-V1.0 * The best open model I…

Paper and discussion history. A mention is not an endorsement.

Reflexion: Language Agents with Verbal Reinforcement Learning — mentioned by krychu

Mon, 10 Jul 2023 07:10:31 GMT

Chain of thought: https://arxiv.org/abs/2201.11903 ReAct: https://arxiv.org/abs/2210.03629 Reflexion: https://arxiv.org/abs/2303.11366 Tree of thoughts: https://arxiv.org/abs/2305.10601 Good video on "Tree of thoughts" which also reviews / puts it in the context of other methods: https://www.youtube.com/watch?v=ut5kp56wW_4 Completion vs conversational interface is something you can read about in the OpenAI API documentation. For the remaining things I don't have single specific pointer at hand.

Paper and discussion history. A mention is not an endorsement.

Reflexion: Language Agents with Verbal Reinforcement Learning — mentioned by famouswaffles

Tue, 27 Jun 2023 21:16:46 GMT

>GPT cannot function in any environment on its own. Except in response to a direct instruction it cannot plan, it cannot take action, it cannot learn, it cannot adjust to changing circumstance. Sure it can. It's not default behavior sure but it's fairly trivial to set up, just expensive. Gpt-4 can loop on its "thoughts" and reflect, it can take actions in the real world. https://tidybot.cs.princeton.edu/ https://arxiv.org/abs/2304.03442 https://arxiv.org/abs/2210.03629 https://arxiv.org/abs/2303.11366

Paper and discussion history. A mention is not an endorsement.

Reflexion: Language Agents with Verbal Reinforcement Learning — mentioned by stevenhuang

Wed, 10 May 2023 17:02:19 GMT

If performance is a concern, just re-prompt GPT if there are any performance concerns, and to propose fixes for them if there are. This is called reflexion and has been found to be an effective way to improve response accuracy and usefulness (also reduces hallucinations). https://arxiv.org/abs/2303.11366

Paper and discussion history. A mention is not an endorsement.

Reflexion: Language Agents with Verbal Reinforcement Learning — mentioned by famouswaffles

Fri, 21 Apr 2023 00:58:10 GMT

>Idk about “above average baseline” That's the minimum. It's far above average for most things. >Maybe, because quite a lot of people aren’t very smart. Lol okay..and ? >But even GPT4 struggles with complex logic problems So does everyone else >and when it makes logic errors it makes ones a human would spot people spot mistakes they're capable of making all the time and so can GPT-4 - https://arxiv.org/abs/2303.11366 https://twitter.com/johnjnay/status/1639362071807549446 >It also has trouble counting letters and words, though that’s largely due to its autoregressive nature. It sees tokens…

Paper and discussion history. A mention is not an endorsement.

Reflexion: Language Agents with Verbal Reinforcement Learning — mentioned by famouswaffles

Tue, 18 Apr 2023 00:47:46 GMT

>the systems operation are well understood That's like saying human behavior is well understood because we know how neurons communicate signals. It's too low level to be useful, hence psychology. >They don't have the ability to reason or reflect. Yes they do https://selfrefine.info/ https://arxiv.org/abs/2303.11366

Paper and discussion history. A mention is not an endorsement.

Reflexion: Language Agents with Verbal Reinforcement Learning — mentioned by famouswaffles

Tue, 11 Apr 2023 02:20:19 GMT

You've been told it doesn't need that much handholding. https://arxiv.org/abs/2303.11366 https://arxiv.org/abs/2303.17651 Why insist otherwise ?

Paper and discussion history. A mention is not an endorsement.

Reflexion: Language Agents with Verbal Reinforcement Learning — mentioned by stevenhuang

Thu, 06 Apr 2023 17:48:26 GMT

That part on providing the opportunity to check its answers is called reflexion and has been shown to significantly improve accuracy and reliability of LLM responses. Very cool. https://arxiv.org/abs/2303.11366 https://evjang.com/2023/03/26/self-reflection.html

Paper and discussion history. A mention is not an endorsement.

Reflexion: Language Agents with Verbal Reinforcement Learning — mentioned by Veen

Mon, 03 Apr 2023 14:34:30 GMT

There's a lot of work happening at the moment around self-reflection and getting LLMs to identify and correct their own hallucinations and mistakes. https://arxiv.org/pdf/2303.11366.pdf

Paper and discussion history. A mention is not an endorsement.

Reflexion: Language Agents with Verbal Reinforcement Learning — mentioned by macrolime

Sat, 01 Apr 2023 14:40:19 GMT

It performs much better when you add reflection to ReAct. https://arxiv.org/pdf/2303.11366.pdf

Paper and discussion history. A mention is not an endorsement.

Reflexion: Language Agents with Verbal Reinforcement Learning — HN submission

Sat, 01 Apr 2023 06:28:48 GMT

Reflexion: An autonomous agent with dynamic memory and self-reflection

2 points; 0 discussion replies on HN. Paper and source history. A submission is not an endorsement.

Reflexion: Language Agents with Verbal Reinforcement Learning — mentioned by kcorbitt

Wed, 29 Mar 2023 00:52:24 GMT

Check out https://arxiv.org/abs/2303.11366?s=09 ;) We aren't introspecting previous runs or human examples to optimize workflows right now, but it's a powerful tool and one that I expect we'll employ in the future to make Taxy more efficient and reliable.

Paper and discussion history. A mention is not an endorsement.

Reflexion: Language Agents with Verbal Reinforcement Learning — mentioned by sacred_numbers

Tue, 28 Mar 2023 04:25:06 GMT

The Reflexion paper (https://arxiv.org/abs/2303.11366) that came out recently shows how this kind of mistake might be overcome. Asking the model to think about the answer after it's generated a first draft greatly improves accuracy. Also, prompt engineering such as copying the generated code, pasting it in a new chat and saying "There's a bug in this code, please find it" can go a long way. There is so much low hanging fruit in harnessing the power of these models that is just being ignored because some even lower hanging fruit (RLHF, system messages, context window size, plugins, etc) is…

Paper and discussion history. A mention is not an endorsement.

Reflexion: Language Agents with Verbal Reinforcement Learning — HN submission

Mon, 27 Mar 2023 15:18:09 GMT

Reflexion: An autonomous agent with dynamic memory and self-reflection

1 point; 0 discussion replies on HN. Paper and source history. A submission is not an endorsement.

Reflexion: Language Agents with Verbal Reinforcement Learning — mentioned by naasking

Sun, 26 Mar 2023 15:48:04 GMT

Reflexion: an autonomous agent with dynamic memory and self-reflection, https://arxiv.org/abs/2303.11366 Add an interpreter as per this article, and we're basically there.

Paper and discussion history. A mention is not an endorsement.

Reflexion: Language Agents with Verbal Reinforcement Learning — HN submission

Sun, 26 Mar 2023 14:27:04 GMT

Reflexion: An autonomous agent with dynamic memory and self-reflection

4 points; 1 discussion reply on HN. Paper and source history. A submission is not an endorsement.

Reflexion: Language Agents with Verbal Reinforcement Learning — mentioned by Dzugaru

Sat, 25 Mar 2023 17:54:46 GMT

It was already trained like this, RLHF [0] is a loop that has ways to feed it the information to "move forward" - update its weights. It was fine-tuned like this. It certainly did ignite something - maybe sparks of AGI, maybe just sparks of pretending to be more liked by humans (hence Human Feedback in RLHF). My guess is that loops by itself (Reflexion [1], etc.) won't change much. ChatGPT massive logs - maybe. [0] https://huggingface.co/blog/rlhf [1] https://arxiv.org/abs/2303.11366

Paper and discussion history. A mention is not an endorsement.

Reflexion: Language Agents with Verbal Reinforcement Learning — mentioned by isaacfrond

Sat, 25 Mar 2023 06:40:48 GMT

the paper: https://arxiv.org/pdf/2303.11366.pdf it reminds me of the thinking-fast vs. thinking-slow dichotomy. Current llms are the thinking fast type. Funnily people’s complains about its errors are reminiscent of this. It answers just to quick and only with its instant response neural net. A thinking slow answer would be more akin to a chain of thought answer. Allowing the llm a more flexible platform than CoT promptin might well be the next step. Of course it would als multiply compute cost. So it might not be in your 20$ subscription

Paper and discussion history. A mention is not an endorsement.

Reflexion: Language Agents with Verbal Reinforcement Learning — HN submission

Sat, 25 Mar 2023 01:26:14 GMT

Reflexion: An autonomous agent with dynamic memory and self-reflection

3 points; 0 discussion replies on HN. Paper and source history. A submission is not an endorsement.